This action might not be possible to undo. Are you sure you want to continue?
https://www.scribd.com/doc/78629775/CommentarytoEurocode
03/25/2015
text
original
E
E
U
U
R
R
O
O
C
C
O
O
D
D
E
E
2
2
C
C
O
O
M
M
M
M
E
E
N
N
T
T
A
A
R
R
Y
Y
E
E
U
U
R
R
O
O
C
C
O
O
D
D
E
E
2
2
C
C
O
O
M
M
M
M
E
E
N
N
T
T
A
A
R
R
Y
Y
COMMENTARY EUROCODE 2
Copyright: European Concrete Platform ASBL, June 2008
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any
means, electronic, mechanical, photocopying, recording or otherwise, without the prior written permission of the European Concrete
Platform ASBL.
Published by the European Concrete Platform ASBL
Editor: JeanPierre Jacobs
8 rue Volta
1050 Brussels
Layout & Printing by
The European Concrete Platform
All information in this document is deemed to be accurate by the European Concrete Platform ASBL at the time of going into press. It
is given in good faith.
Information on European Concrete Platform document does not create any liability for its Members. While the goal is to keep this
information timely and accurate, the European Concrete Platform ASBL cannot guarantee either. If errors are brought to its attention,
they will be corrected.
The opinions reflected in this document are those of the authors and the European Concrete Platform ASBL cannot be held liable for
any view expressed therein.
All advice or information from the European Concrete Platform ASBL is intended for those who will evaluate the significance and
limitations of its contents and take responsibility for its use and application. No liability (including for negligence) for any loss resulting
from such advice or information is accepted.
Readers should note that all European Concrete Platform publications are subject to revision from time to time and therefore ensure
that they are in possession of the latest version.
This publication is based on the publication: "Guida all'uso dell'eurocodice 2" prepared by AICAP; the Italian Association for Reinforced
and Prestressed Concrete, on behalf of the the Italian Cement Organziation AITEC, and on background documents prepared by the
Eurocode 2 Project Teams Members, during the preparation of the EN version of Eurocode 2 (prof A.W. Beeby, prof H. Corres Peiretti,
prof J. Walraven, prof B. Westerberg, prof R.V. Whitman).
Authorization has been received or is pending from organisations or individuals for their specific contributions.
Attributable Foreword to the Commentary and Worked Examples to EC2
Eurocodes are one of the most advanced suite of structural codes in the world. They
embody the collective experience and knowledge of whole of Europe. They are born
out of an ambitious programme initiated by the European Union. With a wealth of
code writing experience in Europe, it was possible to approach the task in a rational
and logical manner. Eurocodes reflect the results of research in material technology
and structural behaviour in the last fifty years and they incorporate all modern trends
in structural design.
Like many current national codes in Europe, Eurocode 2 (EC 2) for concrete
structures draws heavily on the CEB Model Code. And yet the presentation and
terminology, conditioned by the agreed format for Eurocodes, might obscure the
similarities to many national codes. Also EC 2 in common with other Eurocodes,
tends to be general in character and this might present difficulty to some designers at
least initially. The problems of coming to terms with a new set of codes by busy
practising engineers cannot be underestimated. This is the backdrop to the publication
of ‘Commentary and Worked Examples to EC 2’ by Professor Mancini and his
colleagues. Commissioned by CEMBUREAU, BIBM, EFCA and ERMCO this
publication should prove immensely valuable to designers in discovering the
background to many of the code requirements. This publication will assist in building
confidence in the new code, which offers tools for the design of economic and
innovative concrete structures. The publication brings together many of the
documents produced by the Project Team during the development of the code. The
document is rich in theoretical explanations and draws on much recent research.
Comparisons with the ENV stage of EC2 are also provided in a number of cases. The
chapter on EN 1990 (Basis of structural design) is an added bonus and will be
appreciated by practioners. Worked examples further illustrate the application of the
code and should promote understanding.
The commentary will prove an authentic companion to EC 2 and deserves every
success.
Professor R S Narayanan
Chairman CEN/TC 250/SC2 (2002 – 2005)
Foreword to Commentary to Eurocode 2 and Worked Examples
When a new code is made, or an existing code is updated, a number of principles should
be regarded:
1. Codes should be based on clear and scientifically well founded theories,
consistent and coherent, corresponding to a good representation of the structural
behaviour and of the material physics.
2. Codes should be transparent. That means that the writers should be aware, that the
code is not prepared for those who make it, but for those who will use it.
3. New developments should be recognized as much as possible, but not at the cost
of too complex theoretical formulations.
4. A code should be openminded, which means that it cannot be based on one
certain theory, excluding others. Models with different degrees of complexity may
be offered.
5. A code should be simple enough to be handled by practicing engineers without
considerable problems. On the other hand simplicity should not lead to significant
lack of accuracy. Here the word “accuracy” should be well understood. Often so
called “accurate” formulations, derived by scientists, cannot lead to very accurate
results, because the input values can not be estimated with accuracy.
6. A code may have different levels of sophistication. For instance simple, practical
rules can be given, leading to conservative and robust designs. As an alternative
more detailed design rules may be offered, consuming more calculation time, but
resulting in more accurate and economic results.
For writing a Eurocode, like EC2, another important condition applies. International
consensus had to be reached, but not on the cost of significant concessions with regard to
quality. A lot of effort was invested to achieve all those goals.
It is a rule for every project, that it should not be considered as finalized if
implementation has not been taken care of. This book may, further to courses and
trainings on a national and international level, serve as an essential and valuable
contribution to this implementation. It contains extensive background information on the
recommendations and rules found in EC2. It is important that this background
information is well documented and practically available, as such increasing the
transparency. I would like to thank my colleagues of the Project Team, especially Robin
Whittle, Bo Westerberg, Hugo Corres and Konrad Zilch, for helping in getting together
all background information. Also my colleague Giuseppe Mancini and his Italian team
are gratefully acknowledged for providing a set of very illustrative and practical working
examples. Finally I would like to thank CEMBURAU, BIBM, EFCA and ERMCO for
their initiative, support and advice to bring out this publication.
Joost Walraven
Convenor of Project Team for EC2 (1998 2002)
Eurocode 2 Commentary Reading key
Clause 2.1 of the EN 199211
Comment on Clause 2.1
of EN 199211
Comment on Clause 5.8.3
of EN 199211
without “C”: first paragraph of the
comment on clause 5.8.3.1
Guide to EC2 Summary
Page 11 Table of contents
COMMENTARY TO EUROCODE 2  SUMMARY
SECTION 1 SYMBOLS  11
SECTION 2 BASIS OF DESIGN  21
2.1 REQUIREMENTS  21
2.2 PRINCIPLES OF LIMIT STATE DESIGN  25
2.3 BASIC VARIABLES  26
2.4 VERIFICATION BY THE PARTIAL FACTOR METHOD  210
2.5 DESIGN ASSISTED BY TESTING  214
2.6 SUPPLEMENTARY REQUIREMENTS FOR FOUNDATIONS  214
2.7 REQUIREMENTS FOR FASTENINGS  214
SECTION 3 MATERIALS  31
3.1 CONCRETE  31
3.2 REINFORCING STEEL  311
3.3 PRESTRESSING STEEL  312
3.4 PRESTRESSING DEVICES  313
SECTION 4 DURABILITY AND COVER TO REINFORCEMENT  41
SECTION 5 STRUCTURAL ANALYSIS  51
5.1 GENERAL  51
5.2 GEOMETRIC IMPERFECTIONS  51
5.3 IDEALISATION OF THE STRUCTURE 52
5.4 LINEAR ELASTIC ANALYSIS  54
5.5 LINEAR ANALYSIS WITH LIMITED REDISTRIBUTION 54
5.6 PLASTIC ANALYSIS  55
5.7 NONLINEAR ANALYSIS  55
5.8 ANALYSIS OF SECOND ORDER EFFECTS WITH AXIAL LOAD  56
5.9 LATERAL INSTABILITY OF SLENDER BEAMS  534
5.10 PRESTRESSED MEMBERS AND STRUCTURES  534
5.11 ANALYSIS FOR SOME PARTICULAR STRUCTURAL MEMBERS  534
SECTION 6 ULTIMATE LIMIT STATES (ULS)  61
6.1 BENDING WITH OR WITHOUT AXIAL FORCE  61
6.2 SHEAR 612
Guide to EC2 Summary
Page 12 Table of contents
6.3 TORSION  625
6.4 PUNCHING  626
6.5 DESIGN WITH STRUT AND TIE MODELS  632
6.6 ANCHORAGES AND LAPS  632
6.7 PARTIALLY LOADED AREAS  632
6.8 FATIGUE  632
SECTION 7. SERVICEABILITY LIMIT STATES (SLS)  71
7.1 GENERAL  71
7.2 STRESS LIMITATION  71
7.3 CRACK CONTROL  71
7.4 DEFLECTION CONTROL  715
SECTION 8 DETAILING OF REINFORCEMENT AND PRESTRESSING TENDONS  GENERAL  81
8.1 GENERAL  81
8.2 SPACING OF BARS  81
8.3 PERMISSIBLE MANDREL DIAMETERS FOR BENT BARS  81
8.4 ANCHORAGE OF LONGITUDINAL REINFORCEMENT  81
8.5 ANCHORAGE OF LINKS AND SHEAR REINFORCEMENT  81
8.6 ANCHORAGE BY WELDED BARS  81
8.7 LAPS AND MECHANICAL COUPLERS  81
8.8 ADDITIONAL RULES FOR LARGE DIAMETER BARS  81
8.9 BUNDLED BARS  81
8.10 PRESTRESSING TENDONS  81
SECTION 11 LIGHTWEIGHT CONCRETE 111
11.1 GENERAL 111
11.3 MATERIALS  111
Guide to EC2 Section 1
Page 11 Table of contents
SECTION 1 SYMBOLS SECTION 1. SYMBOLS
For the purposes of this document, the following symbols apply.
Note: the notation used is based on ISO 3898:1987
Latin upper case letters
A Accidental action
A Cross sectional area
Ac Cross sectional area of concrete
Ap Area of a prestressing tendon or tendons
As Cross sectional area of reinforcement
As,min minimum cross sectional area of reinforcement
Asw Cross sectional area of shear reinforcement
D Diameter of mandrel
DEd Fatigue damage factor
E Effect of action
Ec, Ec(28) Tangent modulus of elasticity of normal weight concrete at a stress of ıc = 0
and at 28 days
Ec,eff Effective modulus of elasticity of concrete
Ecd Design value of modulus of elasticity of concrete
Ecm Secant modulus of elasticity of concrete
Ec(t) Tangent modulus of elasticity of normal weight concrete at a stress of ıc = 0 and
at time t
Ep Design value of modulus of elasticity of prestressing steel
Es Design value of modulus of elasticity of reinforcing steel
Eǿ Bending stiffness
EQU Static equilibrium
F Action
Fd Design value of an action
Fk Characteristic value of an action
Gk Characteristic permanent action
ǿ Second moment of area of concrete section
L Length
M Bending moment
MEd Design value of the applied internal bending moment
N Axial force
NEd Design value of the applied axial force (tension or compression)
P Prestressing force
P0 Initial force at the active end of the tendon immediately after stressing
Qk Characteristic variable action
Qfat Characteristic fatigue load
R Resistance
S Internal forces and moments
S First moment of area
SLS Serviceability limit state
T Torsional moment
TEd Design value of the applied torsional moment
ULS Ultimate limit state
V Shear force
VEd Design value of the applied shear force
Latin lower case letters
a Distance
a Geometrical data
ǻa Deviation for geometrical data
b Overall width of a crosssection, or actual flange width in a T or L beam
bw Width of the web on T, I or L beams
d Diameter; Depth
d Effective depth of a crosssection
dg Largest nominal maximum aggregate size
e Eccentricity
fc Compressive strength of concrete
Guide to EC2 Section 1
Page 12 Table of contents
fcd Design value of concrete compressive strength
fck Characteristic compressive cylinder strength of concrete at 28 days
fcm Mean value of concrete cylinder compressive strength
fctk Characteristic axial tensile strength of concrete
fctm Mean value of axial tensile strength of concrete
fp Tensile strength of prestressing steel
fpk Characteristic tensile strength of prestressing steel
fp0,1 0,1% proofstress of prestressing steel
fp0,1k Characteristic 0,1% proofstress of prestressing steel
f0,2k Characteristic 0,2% proofstress of reinforcement
ft Tensile strength of reinforcement
ftk Characteristic tensile strength of reinforcement
fy Yield strength of reinforcement
fyd Design yield strength of reinforcement
fyk Characteristic yield strength of reinforcement
fywd Design yield of shear reinforcement
h Height
h Overall depth of a crosssection
i Radius of gyration
k Coefficient; Factor
l (or l or L) Length; Span
m Mass, reduced moment
n reduced axial force
r Radius
1/r Curvature at a particular section
t Thickness
t Time being considered
t0 The age of concrete at the time of loading
u Perimeter of concrete crosssection, having area Ac
u,v,w Components of the displacement of a point
x Neutral axis depth
x,y,z Coordinates
z Lever arm of internal forces
Greek lower case letters
o Angle; ratio
 Angle; ratio; coefficient
¸ Partial factor
¸A Partial factor for accidental actions A
¸C Partial factor for concrete
¸F Partial factor for actions, F
¸F,fat Partial factor for fatigue actions
¸C,fat Partial factor for fatigue of concrete
¸G Partial factor for permanent actions, G
¸M Partial factor for a material property, taking account of uncertainties in the material
property itself, in geometric deviation and in the design model used
¸P Partial factor for actions associated with prestressing, P
¸Q Partial factor for variable actions, Q
¸S Partial factor for reinforcing or prestressing steel
¸S,fat Partial factor for reinforcing or prestressing steel under fatigue loading
¸f Partial factor for actions without taking account of model uncertainties
¸g Partial factor for permanent actions without taking account of model uncertainties
¸m Partial factors for a material property, taking account only of uncertainties in the
material property
į Increment/redistribution ratio
ç Reduction factor/distribution coefficient
İc Compressive strain in the concrete
İc1 Compressive strain in the concrete at the peak stress fc
İcu Ultimate compressive strain in the concrete
İu Strain of reinforcement or prestressing steel at maximum load
İuk Characteristic strain of reinforcement or prestressing steel at maximum load
ș Angle
ì Slenderness ratio
Guide to EC2 Section 1
Page 13 Table of contents
µ Coefficient of friction between the tendons and their ducts
v Poisson's ratio
v Strength reduction factor for concrete cracked in shear
ç Ratio of bond strength of prestressing and reinforcing steel
ȡ Ovendry density of concrete in kg/m
3
ȡ1000 Value of relaxation loss (in %), at 1000 hours after tensioning and at a mean
temperature of 20°C
ȡl Reinforcement ratio for longitudinal reinforcement
ȡw Reinforcement ratio for shear reinforcement
ıc Compressive stress in the concrete
ıcp Compressive stress in the concrete from axial load or prestressing
ıcu Compressive stress in the concrete at the ultimate compressive strain İcu
t Torsional shear stress
 Diameter of a reinforcing bar or of a prestressing duct
n Equivalent diameter of a bundle of reinforcing bars
ĳ(t,t0) Creep coefficient, defining creep between times t and t0, related to elastic
deformation at 28 days
ĳ(,t0) Final value of creep coefficient
¢ Factors defining representative values of variable actions
¢ 0 for combination values
¢ 1 for frequent values
¢ 2 for quasipermanent values
Guide to EC2 Section 2
Page 21 Table of contents
SECTION 2 BASIS OF DESIGN
2.1 Requirements
2.1.2 Reliability management
SECTION 2 BASIS OF DESIGN
C2.1 Requirements
Eurocode 2, Section 2, Part 1.1 states that concrete structures should be designed in accordance
with the general rules of EN1990 and with actions defined in EN 1991. EN 1992 has some
additional requirements.
In particular, the basic requirements of EN1990 Section 2 are deemed to be satisfied for all
concrete structures if limit state design is carried out with the partial factor method in accordance
with EN1990, and if actions are defined in accordance with EN1991, and if combinations of actions
in accordance with EN1990, and finally if resistance, durability and serviceability are dealt with in
accordance with EN1992.
C2.1.2 Reliability management
EC2 points 2.1.2 and 2.1.3 refer to EN1990 section 2 for all rules in relation to reliability, design
working life and quality management measures. These rules, as well as the basic concepts of
structural reliability, will be referred to later.
By the definition given by EN1990, structural reliability is the ability of a structure or a structural
member to fulfil the specified requirements for which it has been designed; it includes structural
safety, serviceability and durability.
Given the random nature of quantities involved in structural design (actions, geometry, restraints,
strength of materials, etc.), the assessment of structural reliability cannot be set up by deterministic
methods, but requires a probabilistic analysis. The objective of safety verification is therefore to
keep failure probability, i.e. probability that a certain danger condition is attained or exceeded,
below a fixed value. This value is determined as a function of type of construction, influence on
safety of people and damage to goods.
Every situation which is dangerous for a construction is referred to as a “limit state”. Once a
construction has attained this condition, it is no longer able to fulfil the functions for which it has
been designed. Limit states are of two types: ultimate limit states and serviceability limit states 
depending on the gravity of their consequences. Exceeding the first causes collapse of the whole
structure or of part of it, exceeding the second causes limited damage that makes the structure unfit
for the requirements of the project. Exceeding serviceability limit states can be reversible or
irreversible: in the first case, no consequences of actions exceeding the specified service
requirements will remain once those actions are removed; in the second, case, some
consequences will remain. For example, a crack width limit state with limited width is a reversible
limit state, whereas one defined by a high width is irreversible (in fact, if the crack width is high,
once the actions are removed the cracks cannot close).
For a given limit state, let us define S and R as two random variables representing respectively
stress and strength. We recall that by ‘stress’ we mean any effect produced in the structural
members by actions applied or by any other effect such as strain, cracking, increase of reinforcing
steel corrosion. ‘Strength’, on the other hand, means the capacity of a structure to respond to a
given stress. A rigorous assessment of structural safety against a relevant limit state can be carried
out by first introducing a safety factor FS, defined as the ratio between strength R and stress S, or
alternatively by a safety margin MS, defined as the difference between R and S:
either
S
F = R/S , or
S
M = R S ,
Both these factors are random variables like R and S. The distribution of Fs or Ms is then
determined on the basis of the statistical distribution of actions, strengths and geometrical
dimensions of the structure, also taking account of the randomness of the structural scheme.
Finally, the probability of failure is related to a fixed reference period of time T through one of the
following expressions:
{ } { }
f S
P = P R / S 1 = P F 1 s s or { } { }
f S
P = P R  S 0 = P M 0 s s .
Pf represents the probability that failure arises, i.e. that the considered limit state is attained or
exceeded at least once during T.
This analysis (known as level 3 method) is very complex. Because of the difficulty of calculation and
of the limitation of available data (data which often fail to give the probabilistic distributions
Guide to EC2 Section 2
Page 22 Table of contents
necessary for calculation), this method is of limited applicability to the design practice.
Alternatively, if only the first and second order moments (averages and standard deviations) of the
random variables R and S, but not their statistical distributions, are known, the probability of failure
can be estimated based on a ȕ index, called the “reliability index”. Assuming that MS is linear, it was
first defined by Cornell as the ratio between the average value ȝM of MS and its standard deviation
ıM:
M M
ȕ = ȝ / ı .
In circumstances where R and S are not correlated (note that in case of normal distributions non
correlation is equivalent to statistical independence), ȕ is expressed as follows:
( )
2 2
R S R S
ȕ = ȝ ȝ / ı +ı , where
R S R S
ȝ ,ȝ ,ı ,ı are the averages and standard deviations of R and S.
This method (known as “level 2” method or “ȕmethod”) does not generally allow assessment of the
probability of failure, with the exception of the particular case where the relation between MS and the
random variables of the problem is linear and the variables have normal distribution. The probability of
failure, i.e. the probability that the safety margin MS assumes nonpositive values, is given by the
distribution function ¢M of MS calculated in 0:
{ } ( )
f S M
P = P M 0 = 0 s u
Introducing m as the normalized variable of the safety margin MS,
S M
M
M  ȝ
m=
ı
the result:
S M M
M = ȝ + m ı substituted in the expression of Pf gives:
{ } { } { } { }
f S M M M M m m
P = P M 0 = P ȝ + m ı 0 = P m ȝ / ı = P m ȕ = (ȕ) = 1 (ȕ) s s s s u u
where ĭm indicates the distribution function of m.
The reliability index may be expressed in geometrical terms. In fact, if we introduce the normalized
strength and stress variables [ ( )
R R
r = R  ȝ / ı and ( )
S S
s = S  ȝ / ı ], the limit condition (MS = 0) is
represented in the r  s plane by a line that divides the plane into a safe region and an unsafe region
(Figure 2.1). The distance from the origin of the axis of this line equals the reliability index (in
circumstances where R and S are not correlated), so the verification of safety is carried out by
assigning a given value to this distance.
Figure 2.1. Geometric representation of the reliability index in the rs plane
The level 2 method is also difficult to apply in practical design because the necessary data are often
not available, so that another method is used: the partial factor method or semiprobabilistic method
(level 1 method).
This method is based on the compliance with a set of rules that ensure the required reliability of the
structure by using “characteristic values” of the problem variables and a series of “safety elements”.
These are represented by partial safety factors, ¸ which cover the uncertainties in actions and
materials, and by additional elements ǻ for uncertainties in geometry, e.g. to allow for the
randomness of cover to reinforcement and therefore of the effective depth of a reinforced concrete
section.
Guide to EC2 Section 2
Page 23 Table of contents
2.1.3 Design working life,
durability and quality
management
This method does not require that the designer has any probabilistic knowledge, because the
probabilistic aspects of the question of safety are already taken into account in the method
calibration process, i.e. in the choice of characteristic values, partial safety factors etc., fixed in the
Standards. The method is based on the following assumptions:
1. strength and stress are independent random variables;
2. characteristic values of strength and stress are fixed as fractiles of given order of the
respective distributions, on the basis of a given probability;
3. other uncertainties are taken into account by transforming characteristic values into design
values, by applying partial safety factors and additional elements;
4. the assessment of safety is positive if the design action effects don't exceed the design
strengths.
It has to be pointed out that the characteristic values of actions are fixed as those values with a
given probability of being exceeded during the service life of the structure only if statistical data are
available. Otherwise, characteristic values are fixed as the nominal values prescribed in standards
or specifications, or as target values, for example in the case of accidental actions such as impacts
from road vehicles, explosions etc.
C2.1.3 Design working life, durability and quality management
Independently from the method used for safety evaluation, a structure can be defined as reliable if
positive safety measures have been provided for all its limit states during the whole design working
life Tu. Tu is defined as the period for which a structure is assumed to be usable for its intended
purpose with anticipated maintenance but without major repair being necessary.
Table 2.1 gives the indicative values of design working life for different types of structures.
Table 2.1. Indicative design working life [Table (2.1)  EN1990]
Design working life
category
Indicative design working life
(years)
Examples
1 10 Temporary structures
(*)
2 10 to 25
Replaceable structural parts, e.g. gantry
girders, bearings
3 15 to 30 Agricultural and similar structures
4 50
Building structures and other common
structures
5 100
Monumental building structures, bridges,
and other civil engineering structures
(*)
Structures or parts of structures that can be dismantled with a view to being reused should not be considered as temporary.
The reliability required for structures [within the scope and field of application of EN1990] shall be
achieved through design in accordance with EN1990 to EN1999 and by appropriate execution and
quality management measures.
EN1990 allows for the choice of different levels of reliability, both for structural resistance and for
serviceability. The choice of the levels of reliability for a particular structure should take account of
the relevant factors, including :
÷ the possible cause and /or mode of reaching a limit state;
÷ the possible consequences of failure in terms of risk to life, injury, potential economical losses;
÷ public aversion to failure;
÷ the expense and procedures necessary to reduce the risk of failure.
The levels of reliability relating to structural resistance and serviceability can be achieved by
suitable combinations of protective measures (e.g. protection against fire, protection against
corrosion, etc.), measures relating to design calculations (e.g. choice of partial factors), measures
relating to quality management, measures aimed to reduce errors in design (project supervision)
and execution of the structure (inspection in phase of execution) and other kinds of measures.
Point (B3.1) of EN1990 Annex B defines three classes based on the consequences of failure or
malfunction of the structure (Table 2.2).
Guide to EC2 Section 2
Page 24 Table of contents
Table 2.2. Consequences classes [Table (B1)  EN1990]
Consequences
Class
Description
Examples of buildings and
civil engineering works
CC3
Serious consequences for loss of
human life, or for economic, social
or environmental concerns
Grandstands, public buildings
where consequences of failure are
high (e.g. a concert hall)
CC2
Moderate consequence for loss of
human life; economic, social or
environmental consequences
considerable
Residential and office buildings,
public buildings where
consequences of failure are
medium (e.g. an office building)
CC1
Low consequence for loss of
human life; economic, social or
environmental consequences
small or negligible
Agricultural buildings where
people do not normally enter (e.g.
storage buildings), greenhouses
Classes CC1 and CC2 correspond to importance classes I, II, whereas CC3 corresponds to classes III and IV, as
from EN1998.1, Table 4.3.
From this table it is apparent that it is the importance of the structure concerned which is the
criterion for classification.
Three reliability classes (RC1, RC2, RC3) may be associated with the three consequence classes
CC1, CC2 and CC3, depending on the reliability index ȕ defined above. Table 2.3 gives values of
ȕ for different values of Pf (remembering that the ȕ index allows the estimation of Pf values only if
the relationship between MS and the random variables of the problem is linear and the variables
have normal distribution).
Table 2.3. Relation between ȕ and Pf [Table C1  EN1990]
Pf 10
1
10
2
10
3
10
4
10
5
10
6
10
7
 1,28 2,32 3,09 3,72 4,27 4,75 5,20
Recommended minimum values for ȕ at ultimate limit states are given in Table 2.4, for reference
periods of 1 year and 50 years.
A design using EN 1990 with the partial factors given in annex A1 and EN 1991 to EN 1999 is
considered generally to lead to a structure with a ȕ value greater than 3,8 for a 50 year reference
period (Table 2.3), i.e. with a rough probability of attaining the ULS in 50 years of 7,2·10
5
(Table
2.3).
Table 2.4. Recommended minimum values of the reliability index ȕ (ultimate limit states)
[Table (B2)EN1990]
Reliability
Class
ȕ minimum values at ULS
1 year reference period 50 years reference period
RC3 5,2 4,3
RC2 4,7 3,8
RC1 4,2 3,3
For the serviceability limit states (irreversible), which are less dangerous and do not concern the safety of
people, the failure probability values for structural elements of Class RC2 are roughly 10
1
(1/10) in 50
years and 10
3
(1/1000) in 1 year; this can be deduced from the relation between ȕ and Pf (Table 2.3) for
the values of ȕ given in the last row of Table 2.5.
Table 2.5. Target values of the reliability index for Class RC2 structural members
[Table (C2)EN1990]
Limit state Target reliability index
1 year reference period 50 years reference period
Ultimate 4,7 3,8
Fatigue  1,5 to 3,8
Serviceability
(irreversible)
2,9 1,5
Design supervision and execution inspection measures being equal, reliability differentiation can be
achieved by distinguishing classes of partial factors ¸F of actions to be used in combination of
actions, and applying multiplication factor KF1, different for each reliability class. Values for KF1 are
given in Table 2.6.
Guide to EC2 Section 2
Page 25 Table of contents
2.2 Principles of limit state
design
Table 2.6. KF1 factor for actions [Table (B3)EN1990]
KF1 factor
for actions
Reliability class
RC1 RC2 RC3
KF1 0,9 1,0 1,1
Measures of design and execution management and quality control are aimed at eliminating failures
due to gross errors, and at ensuring that the design resistance is achieved.
Design supervision and execution inspection levels are given at Tables 2.7 and 2.8, with reference
to each reliability class.
Table 2.7. Design supervision levels (DSL) [Table (B4)EN1990]
Design
supervision
levels
Characteristics
Minimum recommended requirements for checking of
calculations, drawings and specifications
DSL3
relating to RC3
Extended
supervision
Checking performed by an organisation different from
that which has prepared the design
DSL2
relating to RC2
Normal supervision
Checking by different persons than those originally
responsible and in accordance with the procedure of the
organisation.
DSL1
relating to RC1
Normal supervision
Selfchecking: checking performed by the person who
has prepared the design
Table 2.8. Inspection levels (IL) [Table (B5)EN1990]
Inspection levels Characteristics Requirements
IL3 relating toRC3 Extended inspection Third party inspection
IL2 relating to RC2 Normal inspection
Inspection in accordance with the procedures of
the organisation
IL1 relating to RC1 Normal inspection Self inspection
Design with partial factors given in EC2 and with the partial factors given in the EN1990 annexes
results in a structure associated with the RC2 reliability class.
C2.2 Principles of limit state design
Eurocodes adopt the partial factors method, or limit states semiprobabilistic method, as the method
for the verification of structural safety.
EC2 point 2.2 refers to EN1990 Section 3 for limit state design rules. These are here repeated,
partially repeating the description of the three safety verification methods in Par. 2.1.2.
Design for limit states shall be based on the use of structural and load models for relevant limit
states. It shall be verified that no limit state is exceeded when relevant design values for actions,
material and product properties and geometrical data are used in these models.
The verifications shall be carried out for all relevant design situations and load cases.
Two categories are defined by the consequences associated with the attainment of a limit state:
ultimate limit state and serviceability limit state. Verification shall be carried out against both
categories; verification of one of the two categories may be omitted only if it can be proven that it is
satisfied by the verification of the other one.
The ultimate limit states are associated with loss of equilibrium of the whole structure, or failure or
excessive deformation of a structural member and they generally concern safety of people.
For the verification of ultimate limit state design actions shall not exceed the design resistance of
the structure. Table 2.9 shows the ULS classification according to EN1990 [(EN1990 point (6.4.1)].
Table 2.9. ULS classification
Notation Definition
EQU Loss of static equilibrium of the structure or any part of it considered as a
rigid body, where :
÷ minor variations in the value or the spatial distribution of actions from a single
source are significant (e.g. selfweight variations)
÷ the strengths of construction materials or ground are generally not governing
STR Internal failure or excessive deformation of the structure or structural members,
including footings, piles, basement walls, etc., where the strength of construction
materials of the structure governs
GEO Failure or excessive deformation of the ground where the strengths of soil or
rock are significant in providing resistance
FAT Fatigue failure of the structure or structural members.
Guide to EC2 Section 2
Page 26 Table of contents
2.3 Basic variables
2.3.1 Actions and
environmental influences
Serviceability limit states correspond to conditions beyond which specified service requirements for
a structure or structural member are no longer met. Exceeding these limits causes limited damage
but means that the structures do not meet design requirements: functional requirements (not only of
the structure, but also of machines and services), comfort of users, appearance (where the term
“appearance” is concerned with high deformation, extensive cracking, etc.), damage to finishes and
to nonstructural members. Usually the serviceability requirements are agreed for each individual
project.
EN1990 indicates three different types of combinations for serviceability limit states verifications:
characteristic combination (called “rare combination” in the previous versions of Eurocodes),
frequent combination and quasipermanent combination.
The choice of combinations to be taken into account is related to the distinction between reversible
and irreversible limit states: frequent and quasipermanent combinations apply to the first case,
characteristic combinations to the second case.
The definition of relevant limit states for a certain construction requires above all the analysis of the
different situations to which it can be exposed.
The situations chosen for design shall cover all situations that can reasonably occur during the
execution and working life of the structure.
In common cases, design situations are classified as:
persistent design situations, referring to conditions of normal use;
transient situations, referring to temporary conditions of the structure, e.g. during construction
or repair;
accidental situations, involving exceptional conditions of the structure or its exposure,
including fire, explosion, impact, etc.;
seismic situations, where the structure is subjected to a seismic event.
Each design situation is characterized by the presence of several types of actions on the structure.
“Action” means, as EN1990 states, either a set of forces (loads) applied to the structure (direct
actions), or a set of imposed deformations or accelerations caused for example, by temperature
changes, moisture variation, uneven settlement or earthquakes (indirect action).
Actions are classified as:
permanent actions (G), the duration of which is continuous and equal to the design working
life of the structure, or for which the variation in magnitude with time is negligible (e.g. self
weight). Those actions, like prestressing or concrete shrinkage, for which the variation is
always in the same direction (monotonic) until the action attains a certain limit value, are
also permanent actions;
variable actions (Q), divided in variable actions with discrete and regular occurrence in time
(e.g. imposed load of people and lowduration imposed load in general on building floors);
and variable actions characterized by variable and nonmonotonic intensity or direction
(e.g. snow, wind, temperature, waves);
accidental actions (A), which are not easily foreseeable and of low duration (e.g.
explosions, impacts, fire).
Each permanent action with low variability has a single characteristic value Gk. This is the case of actions
due to selfweight: they are generally represented through a nominal value calculated on the basis of the
design drawings (structural and nonstructural member dimensions) and of the average specific gravity of
materials (Gk = Gm).
If a permanent action has relevant uncertainties (coefficient of variation bigger than 10%, where the
coefficient of variation is the ratio between standard deviation and average) and if sufficient
statistical information is available, two characteristic values (upper, Gk,sup, and lower, Gk,inf) should
be used. Gk,sup is the 95% fractile and Gk,inf is the 5% fractile of the statistical distribution for G,
which may be assumed to be Gaussian.
There is a 5% probability that these two values will be exceeded, the probability that the real value
of action is more than Gk,sup or less than Gk,inf is less than 5% (fig. 2.2).
Figure 2.2. Characteristic value of a permanent action: Gk = Gm if the coefficient of variation is negligible; a lower and
an upper characteristic value Gk,inf and Gk,sup are defined if the coefficient of variation is high
Guide to EC2 Section 2
Page 27 Table of contents
Each variable action has four representative values. The main representative value of a variable
action is its characteristic value Qk; the other representative values are, in decreasing order:
the combination value, represented as a product ¢0 Qk,
the frequent value, represented as a product ¢1 Qk,
the quasipermanent value, represented as a product ¢2 Qk.
For simplicity, each of these last three values is defined as a fraction of the characteristic value,
obtained by applying a reducing factor to Qk. In reality, the frequent value and the quasipermanent
value are inherent properties of the variable action, and the ¢1 and ¢2 factors are simply the ratios
between these values and the characteristic value. On the other hand, the ¢0 factor, called the
combination factor, determines the level of intensity of a variable action when this action is taken
into account, in design, simultaneously with another variable action, called “leading variable action”,
which is taken into account by its characteristic value.
The ¢0 factor takes therefore into account the low probability of simultaneous occurrence of the
most unfavourable values of independent variable actions. It is used both for ULS verifications and
for irreversible SLS verifications.
The frequent (¢1 Qk) and the quasipermanent (¢2 Qk) values are used for ULS verifications
including accidental actions and for reversible serviceability limit states.
Values of ¢ factors for buildings are defined in the National Annex. Table 2.10 shows the values
recommended by EN1990.
The characteristic value of a variable action has a defined probability, beforehand accepted, of
being exceeded on the unfavourable side within a fixed reference period. This period is normally
coincident with the design working life of the structure.
For the majority of climatic variable actions, as well as for service loads on building floors, the
characteristic value is based upon the probability of 0,02 of its timevarying part being exceeded
within a reference period of one year. In other words, this is equivalent to a mean return period of
50 years.
Table 2.10. Recommended values of ¢ factors for buildings [Table (A1.1)EN1990]
Imposed loads in buildings, category (see EN 199111) ¢0 ¢1 ¢2
Category A: domestic, residential areas 0,7 0,5 0,3
Category B: office areas 0,7 0,5 0,3
Category C: congregation areas 0,7 0,7 0,6
Category D: shopping areas 0,7 0,7 0,6
Category E: storage areas 1,0 0,9 0,8
Category F: traffic area, vehicle weight 30 kN 0,7 0,7 0,6
Category G: traffic area, 30 kN < vehicle weight 160 kN 0,7 0,5 0,3
Category H: roofs 0 0 0
Snow loads on buildings (see EN 199113)* in Finland, Iceland, Norway,
Sweden and other CEN Member, for sites located at altitude H > 1000 m
a.s.l.
0,7 0,5 0,2
Other CEN Member States, for sites located at altitude H < 1000 m a.s.l. 0,5 0,2 0
Wind load on buildings (see EN 199114) 0,6 0,2 0
Temperature (nonfire) in buildings (see EN 199115) 0,6 0,5 0
NOTE The ¢ values may be set by the National annex.
* For countries not mentioned below, see relevant local conditions.
The characteristic value of seismic action for ULS verification is fixed by Eurocode 8 (EN1998),
based on a return period of 475 years, corresponding to a probability of 10% of being exceeded in
50 years. It is possible to modify the return period by means of an importance factor ¸I.
The frequent value and the quasipermanent value of floor loads on building are determined so that
the average periods of time within which they are exceeded are respectively 10% (ratios of the sum
of intercepts of time BC, DE, FG, HI and the reference period of 50 years represented by segment
AJ in Fig. 2.3) and 50% (ratio of listed segments and segment AJ). Figure 2.3 resumes the
representative values of variable actions.
Guide to EC2 Section 2
Page 28 Table of contents
Figure 2.3. Schematic illustration of representative values of variable actions
For accidental actions, a single nominal value is determined because, due to the nature of these
actions, sufficient information for the appropriate application of statistical methods is not available.
In order to take into account the uncertainties on the choice of characteristic values for actions and
some uncertainties concerning the action modelling, design does not use characteristic values, but
amplified values, called “design values”, which are obtained by multiplying characteristic values by a
partial factor.
Symbols representing the design values are indicated with index d. Table 2.11 shows the steps to
pass from the representative values of actions to the design values of their effects on construction.
Table 2.11. Procedure to determine the design values of effects on structures
starting from the representative values of actions
Expression Comment
i
F
Actions on the structure are identified
k,i
F o
( )
k,i
F ¢
where
( )
0 1 2
ȥ ȥ , ȥ , ȥ =
Representative values are assigned to actions: characteristic
values or other (combination, frequent, quasipermanent)
values.
Ȗ
d,i f,i k,i
F F =
( )
o Ȗ ȥ
f,i k,i
F where
( )
0 1 2
ȥ ȥ , ȥ , ȥ =
Design values of actions are determined by multiplying the
representative values
k,i
F or ȥ
k,i
F (where 0 1
ȥ ȥ , ȥ , ȥ =
2 ) by a
partial factor Ȗ
f,i
. Ȗ
f,i
is a partial factor generally covering the
uncertainties related to the choice of characteristic values for
actions and, sometimes, part of the uncertainties related to
action modelling. In case of permanent actions, when it is
necessary to split the action into a favourable and an
unfavourable part, two different partial factors, indicated as
Ȗ
G,sup
and Ȗ
G,inf
, are used.
( )
Ȗ ȥ
d f,i k,i d
E = E F ; a
Actions that can occur simultaneously are considered;
combinations of actions are calculated and the effects of these
combinations on the structure are assessed (e.g. action effect
in a cross section).
d
a represents either the design value of the set of geometrical
data (in general, values indicated on the design drawings) or
data that take into account the possibility of geometrical
imperfection liable to cause second order effects.
( )
Ȗ Ȗ ȥ
d Sd f,i k,i d
E = E F ; a
The design value of effects is obtained by multiplying the
values produced by the design actions, by a partial factor
Sd
Ȗ
mainly covering the uncertainties of the structural model.
( )
Ȗ ȥ
d F,i k,i d
E = E F ; a
In normal cases, the previous expression is simplified in this
one, where: Ȗ (Ȗ Ȗ )
F,i Sd f,i
= f , so that the model coefficient Ȗ
Sd
does not explicitly appear.
The product: Ȗ
d,i F,i k,i
F = F or
( )
0 1 2
Ȗ ȥ ȥ ȥ , ȥ , ȥ
F,i k,i
F ; = is
often directly assumed as the design value of the action Fk,i.
Guide to EC2 Section 1
Page 14 Table of contents
Guide to EC2 Section 2
Page 29 Table of contents
2.3.1.2 Thermal effects
2.3.1.3 Differential
settlements/movements
2.3.1.4 Prestress
2.3.2 Material and product
properties
2.3.2.1 General
Several material properties are involved in structural design. The main one is strength, i.e. the ability to
resist forces without breaking or failing.
Strength of materials is represented through a characteristic value, indicated as fk. This is the value that
has a given probability of not being attained or exceeded during a hypothetical unlimited test series.
Eurocode EN1990 defines the characteristic value of a property of a material as the 5% fractile of its
statistical distribution where a minimum value of the property is the nominal failure limit (general
case), and as the 95% fractile where a maximum value is the limiting value.
For the structural stiffness parameters (moduli of elasticity, creep coefficient, thermal expansion
coefficient, etc.), the characteristic value is taken as a mean value because, depending on the case,
these parameters can be favourable or unfavourable.
Product properties are also represented by a single characteristic value or a set of characteristic
values, according to their constituent materials.
Table 2.12 shows the steps to pass from the characteristic values of individual material strengths or
product resistances to the design values of structural resistance.
Table 2.13 gives the values of partial safety factors to be assumed for concrete and steel for ULS,
in case of persistent, transient and accidental load combinations.
Table 2.12. Procedure to determine the design values of resistances starting from the characteristic values of strength
Expression Comment
i
X
Material strengths and product resistances involved in the
verifications are identified.
k, i
X
Characteristic values of material strengths and product resistances
are introduced.
k,i
d,i
m,i
X
X Ș
Ȗ
=
The design value of a material property is determined on the basis of its
characteristic value, through the two following operations:
a) divide by a partial factor Ȗ
m
, to take into account unfavourable
uncertainties on the characteristic of this property, as well as any
local defaults.;
b) multiply, if applicable, by a conversion factor q mainly aimed at
taking into account scale effects.
 
q


¸
\ .
k, i
d
m, i
X
R ; a
Determine the structural resistance on the basis of design values of
individual material properties and of geometrical data.
k,i
d d
Rd m,i
X 1
R R Ș ; a
Ȗ Ȗ
 
=


\ .
Following a procedure similar to the one for calculating the design
value of action effects, the design value of structural resistance is
determined on the basis of individual material properties and of
geometrical data multiplied by a partial factor Rd
Ȗ
that covers the
model uncertainties of resistance and the geometrical data
variations, if these are not explicitly taken into account in the model.
k,i
d d
M,i
X
R R Ș ; a
Ȗ
 
=


\ .
As for the action effects, factor Rd
Ȗ
is often integrated in the global
safety factor ¸M,i, by which the characteristic material strength is
divided: M,i Rd m,i
Ȗ f(Ȗ , Ȗ ) =
.
Fig. 2.4 summarise in a schematic way the relation between the single partial factors used in
Eurocodes.
Figure 2.4. Relation between the single partial factors [Fig. (C3)  EN1990]
Guide to EC2 Section 2
Page 210 Table of contents
2.3.2.2 Shrinkage and creep
.
2.3.3 Deformations of
concrete
2.3.4 Geometric data
2.3.4.1 General
2.3.4.2 Supplementary
requirements for cast in place
piles
2.4 Verification by the partial
factor method
2.4.1 General
2.4.2 Design values
2.4.2.1 Partial factor for
shrinkage action
2.4.2.2 Partial factors for
prestress
2.4.2.3 Partial factor for
fatigue loads
2.4.2.4 Partial factors for
materials
Table 2.13. Partial factors for concrete and steel for ultimate limit states [Table (2.1N)EC2]
Design situations ¸C for concrete ¸S for reinforcing
steel
¸S for prestressing
steel
Persistent and transient 1,5 1,15 1,15
Accidental 1,2 1,0 1,0
The values of partial factors given in the previous table were determined as:
( )
M R f
= exp 3,04V  1,64V ¸
where:
( )
2 2 2
R m G f
V = V + V + V
VR coefficient of variation of resistance
Vm coefficient of variation of model uncertainty
VG coefficient of variation of geometrical factor
Vf coefficient of variation of material strength
The basic values in EC2 may be considered to be based on the following assumptions:
For reinforcement model uncertainty Vm = 2,5 %
geometry VG = 5 %
steel strength Vf = 4 %
For these values the equation gives ¸M = ¸s = 1,154
For concrete Variation of model uncertainty 5 %
Variation of geometry 5 %
Variation of concrete strength 15 %
For these values the equation gives ¸M = ¸c = 1,30. Assuming an additional factor 1,15 to cover the
uncertainty arising from the concrete being tested with test specimens specially made and cured for
this purpose, rather than from the finished structure, the result is ¸c = 1,15·1,30 = 1,50.
For serviceability limit states values of partial safety factors ¸C and ¸S are defined in the National
Annex. The recommended value for situations not covered by specific parts of EC2 is 1,0.
Besides the partial safety factors for materials above, EC2 also defines partial factors for shrinkage,
prestressing, fatigue loads and materials for foundations. These values are also defined in the
National Annex. The table below gives the values recommended in EC2.
Guide to EC2 Section 2
Page 211 Table of contents
2.4.2.5 Partial factors for
materials for foundations
2.4.3 Combinations of actions
Table 2.14. Recommended values of partial safety factors
Shrinkage (e.g. for ULS verification of stability when second order effects are
relevant)
¸SH
1,0
Prestressing
favourable in persistent and transient situations ¸P,fav
1,0
Stability ULS with external prestressing if an increase of
prestressing may be unfavourable
¸P,unfav
1,3
local effects ¸P,unfav
1,2
Fatigue ¸F,fat
1,0
Materials for foundations (amplified in order to obtain the design resistance of
cast in place piles without permanent scaffolding)
¸C,fond 1,1 ¸C
The general formats for the ULS and SLS combinations of actions, as defined in EN 1990 Section
6, are given below.
It has to be pointed out that the combinations foreseen in EN 1990 cover static actions only, but
there is no consideration of actions of dynamic type, such as actions caused by vibrating machinery
(turbines, compressors, etc.).
It has also to be noted that EN 1990 considers the zero value a possible value of the partial factor of
variable actions ¸Q, although ¸Q = 0 has no probabilistic meaning. A zero value of ¸Q is therefore an
expedient to remove from the combination of actions the favourable effect of variable actions (e.g.
to maximize the positive bending moment in the central span of a threespan continuous beam, the
variable imposed load shall be applied on the central span only, which is equivalent to setting ¸Q = 0
in the two lateral spans, see Example 2.1).
ULS Combinations
Three types of combinations of actions should be considered for Ultimate Limit States: fundamental,
accidental and seismic.
Combinations of actions for persistent or transient design situations (fundamental combinations):
G, j k, j P Q,1 k,1 Q,i 0,i k,i
j 1 i 1
Ȗ G Ȗ P Ȗ Q Ȗ ȥ Q
> >
+ + +
_ _
or alternatively for the STR and GEO limit states, the more unfavourable of the two following
expressions:
G, j k, j P Q,1 0,1 k,1 Q,i 0,i k,i
j 1 i 1
Ȗ G Ȗ P Ȗ ȥ Q Ȗ ȥ Q
> >
+ + +
_ _
j G, j k, j P Q,1 k,1 Q,i 0,i k,i
j 1 i 1
ȟ Ȗ G Ȗ P Ȗ Q Ȗ ȥ Q
> >
+ + +
_ _
where i
ȟ
are reduction factors for the unfavourable permanent actions G.
EN1990 introduces then the possibility to choose for the fundamental combination:
1. the traditional expression with a predominant variable action Qk,1 and the permanent
actions (including prestressing P) introduced with its characteristic value, other concomitant
variable actions introduced with their combination values;
2. or a system of two expressions, the most unfavourable of which should be adopted by the
designer; these expressions are obtained by reducing the multiplying factor of the
predominant variable action (through the ¢0 factor) in the first case, or by reducing the
multiplying factors for permanent unfavourable actions (through the ç factor that is in the
range 0,85 ÷ 1,00).
The expression to be used for the fundamental combination is chosen in the National Annex.
Combinations of actions for accidental design situations
k, j d 1,1 2,1 k,1 2,i k,i
j 1 i 1
G P A (ȥ ȥ )Q ȥ Q or
> >
+ + + +
_ _
where Ad is the accidental design action (for fire situations, it represents the design value of the
indirect thermal action due to fire).
Guide to EC2 Section 2
Page 212 Table of contents
If a variable action may be present on the structure at the moment when the accidental action
occurs, its frequent value (¢1,1 Qk,1) will be used in the combination, otherwise its quasipermanent
value (¢2,1 Qk,1) will be used; other variable actions are introduced in the combination with their
quasipermanent values (¢2,i Qk,i).
Combinations of actions for seismic design situations
k, j Ed 2,i k,i
j 1 i 1
G P A ȥ Q
> >
+ + +
_ _
where AEd is the design value of seismic action ( Ȗ
Ed I Ek
A = A with ¸I importance factor and AEk
characteristic values of seismic action  see EN1998). Note that the seismic action is combined with
the quasipermanent value of variable actions, whereas permanent actions Gk,j are taken into
account with their characteristic value, and prestressing with its representative value (see ch. 6).
SLS combinations
There are three combinations of actions for SLS: characteristic, frequent and quasipermanent.
Characteristic combination
k, j k,1 0,i k,i
j 1 i 1
G P Q ȥ Q
> >
+ + +
_ _
Frequent combination:
k, j 1,1 k,1 2,i k,i
j 1 i 1
G P ȥ Q ȥ Q
> >
+ + +
_ _
Quasipermanent combination:
k, j 2,i k,i
j 1 i 1
G P ȥ Q
> >
+ +
_ _
Detailed expressions of the combinations of actions are given in the normative annexes to EN 1990
(Annex A1 for buildings, A2 for bridges, etc.), together with the recommended values of partial
safety factors ¸F and of combination factors ¢.
The characteristic combination should be normally considered for short term limit states, associated
with the onetime attainment of a fixed value of the effect considered (e.g. crack formation). It
corresponds to those effects that have a probability of being exceeded which is close to the
probability that the characteristic value of the predominant variable action Qk,1 will be exceeded. In
other words, the characteristic combination should be considered for verification of irreversible
serviceability limit states: e.g. the crack width limit state characterized by a 0,5 mm crack width is an
irreversible limit state, because such a wide crack cannot completely close once the action that
produced it is removed.
The frequent combination should be considered for long term limit states, associated with the
attainment of a fixed value of the effect considered for a small fraction of the reference period or for
its attainment a fixed number of times. It corresponds to those effects that are exceeded with a
frequency or length of time close to the one of the frequent value ¢1 Qk,1 of the predominant
variable action (e.g. the crack width limit state of a prestressed concrete beam with bonded
tendons, XC2 exposure class, with design crack width not exceeding 0,2 mm).
The quasipermanent combination should be considered for long term action effects, associated
with the attainment of a fixed value of these limit states for a long fraction of the reference period
(e.g. the crack width limit state for a reinforced concrete or prestressed concrete beam with
unbonded tendons, with design crack width not exceeding 0,3 mm).
Frequent and quasipermanent combinations must be considered for the verification of reversible
limit states, i.e. limit states that will not be attained or exceeded once the actions that have caused
attainment or exceeding are removed.
Combination of actions for fatigue limit state
The combination of actions for fatigue verification is given in [(6.8.3)EC2]:
k, j 1,1 k,1 2,i k,i fat
j 1 i 1
G P ȥ Q ȥ Q Q
> >
 
+ + + +

\ .
_ _
where
fat
Q is the relevant fatigue load (e.g. traffic load or other cyclic load).
Guide to EC2 Section 2
Page 213 Table of contents
2.6.1 Combinations of actions for the ultimate limit states verification of a building
EN1990 Annex A1 gives rules for combinations of actions for buildings, on the basis of symbolic
expressions and recommended values (or of values given in the National Annex) of partial factors to
be applied to actions in the combinations. Eurocodes allow combinations of actions to contain two
or more variable actions.
In general, for ultimate limit states, values of partial factors are subdivided in three sets (A, B and
C), given in Tables [A1.2(A)EN1990] to [A1.2(C)EN1990], which are combined in the following
table.
Table 2.15. Sets A, B and C of partial factors ¸ for actions
Actions
Permanent
actions Gk
Predominant
variable action
Qk,1 (Note 2)
Non predominant
variable actions
Qk,i (Note 2) Unfavourable Favourable
Set A 1,10 0,90 1,5 1,5 · ¢
0,i
Set B
(Note 1)
(eq.6.10)
EN1990
1,35 1,00 1,5 1,5 · ¢
0,i
or alternatively the most unfavourable between the two following:
(eq.6.10a)
EN1990
1,35 1,00 1,5 · ¢
0,1 1,5 · ¢
0,i
(eq.6.10b)
EN1990
0,85 · 1,35 1,00 1,5 1,5 · ¢
0,i
Set C 1,00 1,00 1,30 1,30
Note 1: eq.6.10 or alternatively the most unfavourable between eq.6.10a and eq.6.10b is used. The choice is made in the National Annex. In case
of 6.10a and 6.10b, the National annex may in addition modify 6.10a to include permanent actions only.
Note 2: The partial factor of favourable variable actions should be taken as 0.
Depending on the limit state under consideration, values from one or more sets should be used, as
indicated in the following table.
Table 2.16. Sets of partial factors to be used for ULS
Limit state Set of partial factors to be used
EQU – Static equilibrium Set A
STR – Design of structural members not
involving geotechnical actions
Set B
STR – Design of structural members
involving geotechnical actions
(foundations, piles, basement walls,
etc.)
GEO – Breaking or excessive
deformation of ground
Approach 1(*): Apply in separate calculations design
values from Set B and Set C to all actions. In
common cases, the sizing of foundations is governed
by Set C and the structural resistance is governed by
Approach 2: Apply Set B to all actions
Approach 3: Simultaneously apply, in the same
calculation, Set B to actions on the structure and Set
C to geotechnical actions
(*)
The choice of approach to be used for STR/GEO verification is given in the National Annex.
Combinations obtained with sets A, B and C of partial factors with EN 1990 recommended values
are given below. Note that the partial factor of variable actions should be taken as 0 where these
actions are favourable.
Combinations of actions with Set A of partial factors (EQU)
( )
kj,sup kj,inf k,1 0,i k,i
j 1 i 1
1,1 G 0, 9 G 1, 5 Q 1, 5 ȥ Q
> >
+ + +
_ _
In this combination the favourable part of a same action is multiplied by 0,9 and the unfavourable by
1,1. For example, in the verification of holding down devices for the uplift of bearings of a
continuous beam, the self weight of spans that give a stabilising effect should be multiplied by 0,9
whereas the self weight of spans that give destabilising effect should be multiplied by 1,1
(see Example 2.1).
Guide to EC2 Section 2
Page 214 Table of contents
2.4.4 Verification of static
equilibrium  EQU
2.5 Design assisted by testing
2.6 Supplementary
requirements for foundations
2.7 Requirements for
fastenings
Combinations of actions with Set B of partial factors (STR/GEO)
Either
( )
kj,sup kj,inf k,1 0,i k,i
j 1 i 1
1, 35 G 1, 0 G 1, 5 Q 1, 5 ȥ Q
> >
+ + +
_ _
[eq. (6.10)EN1990]
(where
kj,sup
G are unfavourable permanent loads and
kj,inf
G are favourable permanent loads)
or the less favourable of the two following expressions:
( )
kj,sup kj,inf 01 k,1 0,i k,i
j 1 i 1
1, 35 G 1, 0 G 1, 5 ȥ Q 1, 5 ȥ Q
> >
+ + +
_ _
[eq. (6.10a)EN1990]
( )
kj,sup kj,inf k,1 0,i k,i
j 1 i 1
1,15 G 1, 0 G 1, 5 Q 1, 5 ȥ Q
> >
+ + +
_ _
[eq. (6.10b)EN1990]
The National Annex decides whether eq. [(6.10)EN1990] or the less favourable of [(6.10a)
EN1990] and [(6.10b)EN1990] should be considered.
In these expressions Gkj,sup (Gkj,inf) is a set of permanent actions from one source with an unfavourable
(or favourable) resulting effect of the total action. For example, all actions originating from the self weight
of the structure may be considered as coming from one source; this also applies if different materials are
involved. Therefore in the verifications of resistance of the sections of a beam, its self weight should be
taken with the same design value for the whole length of the beam ( G = 1,35), whereas a different value
of partial factor (¸G = 1,0) can be taken for permanent loads originating from a different source.
Combinations of actions with Set C of partial factors (STR/GEO)
( )
kj,sup kj,inf k,1 0,i k,i
j 1 i 1
1, 0 G 1, 0 G 1, 3 Q 1, 3 ȥ Q
> >
+ + +
_ _
See examples 2.1, 2.2, 2.3, 2.4, 2.5
Guide to EC2 Section 3
Page 31 Table of contents
SECTION 3 MATERIALS
3.1 Concrete
3.1.1 General
3.1.2 Strength
SECTION 3. MATERIALS
Section 3 of Eurocode 2, dedicated to materials, is structured in the following paragraphs:
3.1 – Concrete
3.2 – Reinforcing steel. Annex CEC2 is related to this paragraph
3.3 – Prestressing steel
C3.1.1 General
This section deals with normal weight concrete, viz. according to EN 2061 having density greater
than 2000 but not exceeding 2600 kg/m
3
. “Lightweight” concrete is dealt with in Sect.11–EC2.
C3.1.2 Strength
Compressive strength is defined in [3.1.2(1)PEC2], in accordance with EN 2061, by the
characteristic value fck (5% fractile of distribution) obtained through the elaboration of compression
tests executed at 28 days on cylindrical specimens of diameter 150 mm and height 150 mm.
As in many countries testing is carried out on 150 mm cubic specimens, EN 2061 admits fck,cube
compressive strength, too.
Compressive strength classes are denoted by letter C followed by two numbers that indicate the
cylinder and cube characteristic strength, expressed in N/mm
2
, for example C30/37.
EC2 contemplates 14 classes: from C12/15 to C90/105.
Table [3.1EC2] gives the numeric values of strength and deformation characteristics associated with
strength classes and the analytic relationship expressing such values in function of fck. Average
values of compressive strength fcm , of tensile strength fctm and of elasticity modulus Ecm are plotted in
Fig. 3.1 in function of fck. Ecm is denoted by the inclination of the line secant of the ıİ relation
between points ı = 0 and ı = 0,4·fcm as indicated in [Fig. 3.2EC2].
Clause [3.1.2(6)EC2] deals with the development of compressive strength with time. Formula [(3.2)
EC2] allows to calculate the average strength fcm at a time t (days) in function of the value at 28 days,
which can be deduced from Table [3.1EC2] and from the class of cement used. Cement classes
conforming to EN 197 are:
Class R (rapid hardening), including CEM 42,5R, CEM 52,5N and CEM 52,5R
Class N (normal hardening) including CEM 42,5N and CEM 32,5R
Class S (slow hardening) including CEM 32,5N.
Table 2.1 shows the development until 360 days of average strength fcm of concrete produced with
cement from the three classes, where the compressive strength at 28 days for each class, fcm,
equals 1.
Figure 3.1. Values of fcm , fctm and Ecm in function of fck
Table 3.1. Development of fcm(t) / fcm,28 ratio
t (days) Class R Class N Class S
1 0,42 0,34 0,19
2 0,58 0,50 0,35
3 0,66 0,60 0,46
7 0,82 0,78 0,68
14 0,92 0,90 0,85
28 1,00 1,00 1,00
90 1,09 1,11 1,18
360 1,15 1,19 1,31
Guide to EC2 Section 3
Page 32 Table of contents
3.1.3 Elastic deformation
3.1.4 Creep and shrinkage
3.1.4 (15)
3.1.4 (6)
C3.1.3 Elastic deformation
Clause [3.1.3EC2] deals with the elastic deformation represented by the values of the modulus of
elasticity Ecm . Note that the numerical values in Table [3.1EC2] and here given in Fig 3.1 are referred
to concrete produced with siliceous aggregate, at 28 days of curing. These values should be reduced,
for concrete with limestone and sandstone aggregates, respectively by 10% and by 30%. For basaltic
aggregates they should be increased by 10%.
The development of the Emodulus with time is deduced from the one of fcm (Table 3.1) with formula
[(3.5)EC2]. Table 3.2 shows the development of Ecm up to 360 days as a ratio of Ecm at 28 days.
Table 3.2. Development of modulus of elasticity Ecm with time
t (days) Class R Class N Class S
3 0,88 0,86 0,80
7 0,94 0,93 0,89
14 0,97 0,96 0,95
28 1,00 1,00 1,00
90 1,02 1,03 1,05
360 1,04 1,05 1,08
C3.1.4 Creep and shrinkage
Clause [3.1.4EC2] is about creep and shrinkage.
These two phenomena are typical of concrete. The first relates to the increase in the deformation with
time in presence of permanent actions, the second is a spontaneous variation of volume. The
development of both phenomena depends on the ambient humidity, the dimensions of the element
and the composition of the concrete. Creep is also influenced by the maturity of the concrete when the
load is first applied and depends on the duration and magnitude of the loading.
C3.1.4 (15)
The creep deformation of concrete İcc(·,to) at time t = for a constant compressive stress
ıc applied at the concrete age t0, is given by:
( ) ( )
  o
c · = ¢ ·

\ .
c
cc 0 0
c
, t , t
E
[(3.6)EC2] (2.1)
where ¢(·, t0) is the creep coefficient related to Ec , the tangent modulus, which may be taken as
1,05 Ecm as from Table [3.1EC2]. Annex B of the Eurocode gives detailed information on the
development of creep with time. Where great accuracy is not required, the value found from Figure
3.1 may be considered as the creep coefficient, provided that the concrete is not subjected to a
compressive stress greater than 0,45fck(t0) at an age t0.
The values given in Figure 3.1 are valid for ambient temperatures between 40°C and +40°C and a
mean relative humidity between RH = 40% and RH = 100%. Moreover, graphs are in function of the
concrete t0, expressed in days, at the time of loading, of the notional size ho = 2Ac/u where Ac is the
concrete crosssectional area and u is the perimeter of that part which is exposed to drying. They are
also in function of the concrete class (e.g. C30/37) and of the class (R, N, S) of cement used, as
detailed at clause [3.1.2(6)EC2].
When the compressive stress of concrete at an age t0 exceeds the value 0,45 fck(t0) then the
proportionality expressed by [(3.1)EC2] does not subsist and creep nonlinearity should be
considered. In such cases the nonlinear notional creep coefficient should be obtained as from the
exponential expression [(3.7)EC2].
C3.1.4 (6)
The total shrinkage strain İcs is composed of two components:
İcd, the drying shrinkage strain, which develops slowly, since it is a function of the migration of the water
through the hardened concrete and
İca, the autogenous shrinkage strain, which develops during hardening of the concrete: the major part of it
therefore develops in the early days after casting.
Autogenous shrinkage can be defined as “the macroscopic volume reduction of cementitious materials
when cement hydrates after initial setting. Autogenous shrinkage does not include the volume change due
to loss or ingress of substances, temperature variation, the application of an external force and restraint”
[JCI, 1998]. Autogenous shrinkage specially has to be regarded for higher strength concrete’s, since its
value increases with decreasing water cement ratio. Autogenous shrinkage is negligible, in comparison to
drying shrinkage, in concrete having a w/c ratio greater than 0.45, but it can represent 50% of the total
shrinkage when w/c is 0.30. Its development in time is linked to the hardening process of the concrete. In
high strength concrete there is a considerable strength development during the first days; therefore
autogenous shrinkage specially has to be regarded in cases that imposed deformations can occur, such
as in the case that new concrete is cast against old concrete. In Annex B of EC2 the basic equations for
both drying shrinkage and autogenous shrinkage are given. They are valid up to a concrete strength class
C90.
Guide to EC2 Section 3
Page 33 Table of contents
3.1.5 Stressstrain relation for
nonlinear structural analysis
Drying shrinkage is essentially a function of the ambient humidity and of the notional size ho = 2Ac/u.
Clause [3.1.4(6)EC2] gives formulae and tabled values normally used. Further information is given in
Annex B (part B2).
10
0
10
1
10
2
10
3
10
4
10
5
10
6
800
700
600
500
400
300
200
100
0
Shrinkage according to EN 199211
42,5R tcure=1d  RH=65% h
0
=150mm
Age [days]
S
h
r
i
n
k
a
g
e
[
µ
S
]
C12/15
C20/25
C25/30
C30/37
C35/45
C40/50
C45/55
C50/60
C70/85
C90/105
C90/105
C12/15
Figure 3.2. Development of shrinkage according to EC2
C3.1.5 Stressstrain relation for nonlinear structural analysis
Clause [3.1.5EC2] gives the stressstrain relation for nonlinear structural analysis as described by
[Fig. 3.2EC2] and by the expression [(3.14)EC2].
In the ENV199211 the following relation has been used in order to describe the mean stress strain
relation:
( )
2
c
ı 
=
fcm 1+  2
kq q
k q
(1)
where
Ș = İc / İc1 (2)
İc1‰ = 0.0022 (strain at peak compressive stress)
k = (1.1 Ec) İc1/fc (3)
Ec denotes the mean value Ecm of the longitudinal modulus of deformation, where
Ecm = (fck + 8)
1/3
(4)
In ENV 199211, however, only concrete strength classes up to C50/60 were considered.
High strength concrete is known to behave in a more brittle way and the formulation therefore cannot
be extended to high strength concrete without modification.
Fig. 3.2 shows compressive stress–strain relations for concrete strength classes ranging from about
C25 to C90
Figure 3.2 Stressstrain relation for concrete’s different strength classes subjected to a constant strain rate
(strain in horizontal axis in ‰, stress in vertical axis in MPa)
Guide to EC2 Section 3
Page 34 Table of contents
In [CEB, 1995], the following modifications have been proposed:
Eq. (4) overestimates the Emodulus for HSC. An appropriate formulation for HSC is:
Ecm = 22.000 [(fck + ǻf) / 10]
0.3
(5)
where the difference between mean and characteristic strength ǻf is 8 MPa. This equation is also
a good approximation for the Emodulus of normal strength concrete and could therefore be
attributed general validity. (It should be noted that the given values are mean values and that the
real modulus of elasticity can considerably be influenced by a component like the aggregate. If
the modulus of elasticity is important and results from similar types of concrete are not known,
testing of the concrete considered is recommended).
To determine the ascending branch, using Eq. 1, the constant value İc1 =  0.0022 should for HSC
be replaced by
İc1 (‰)=  0.7 fcm
0.31
(fcm in MPa) (6)
This value can as well be used for normal strength concrete
For HSC (>C50) the descending branch should be formulated by
ıc = fcm / [1 + {(Ș1  1) / (Ș2 – 1)}
2
] (7)
where Ș1 = İc / İc1 and Ș2 = (İc1 + İc0) / İc1 where İc0 is a value to be taken from Table 3.3.
Table 3.3.The parameter t for HSC
fck (MPa) 50 60 70 80 90
cc0 (10
3
) 0.807 0.579 0.338 0.221 0.070
A warning is given that the descending branch highly depends on the testing procedure and its
formulation should be used with caution.
Using the relations (1,2,3,5,6,7) the diagram shown in Fig. 3.3 is obtained.
Since the descending branch for HSC is not very reliable, a simplified formulation is preferred, in that
the lines according to Eq. (1) are continued beyond the top, Eq. 6, until a defined value İcu is reached,
according to
İcu ( ‰) = 2.8 + 27 [(98 – fcm)/100]
4
for fck 55 Mpa (8)
In this way the simplified curves shown in Fig. 3.4 may be obtained.
Figure 3.3. Mean stressstrain relations, obtained by combining Eq. (13) and (57).
Guide to EC2 Section 3
Page 35 Table of contents
3.1.6 Design compressive and
tensile strengths
Figure 3.4. Simplified mean stressstrain relations, according to the new formulation, combining Eq. (5,6,8)
C3.1.6 Design compressive and tensile strengths
The value of the design compressive strength fcd is defined as
fcd = occ fck /¸C [(3.15)EC2] (2.2)
where
Įcc is the coefficient taking account of long term effects on the compressive strength and of
unfavourable effects resulting from the way the load is applied;
ȖC is the partial safety factor for concrete, which is 1,50 [Table 2.1NEC2].
A well known research program focussing on the effects of long term loading was the one carried out
by Rüsch [Rüsch, 1960]. He carried out tests on concrete prisms, which he loaded to a certain fraction
of the shortterm compressive strength: subsequently the load was kept constant for a long period. If
the longterm loads were higher than about 80% of the shortterm bearing capacity, failure occurred
after a certain period. Fig. 3.5 reproduces Rüsch’s diagrammatic representation of concrete strains as
a function of the applied stresses for several loading times.
Figure 3.5. Stressstrain relations for several time durations of axial compressive loads (Rüsch, 1960)
As can be seen, the longer the loading time, the more the ultimate strength approaches the longterm
value 80%. The tests carried out by Rüsch were limited to concrete’s with a maximum cube strength
of about 60Mpa. Tests by Walraven and Han on concrete’s with cube strength’s up to 100 Mpa
showed that the sustained loading behaviour for high strength concrete is similar to that of
conventional concrete’s [Han/Walraven, 1993].
Guide to EC2 Section 3
Page 36 Table of contents
3.1.7 Stressstrain relations
for the design of cross
sections
However, Rüsch’s tests were carried out on concrete which had an age of 28 days at the time the
load was applied. This condition will normally not hold for a structure in practice, which generally will
be much older when subjected to a load. This means that the sustained loading effect is at least
partially compensated by the increase in strength between 28 days and the age of loading. Fig. 3.6
shows the strength development in time according to eq. EC3.3 for concrete’s made with rapid
hardening high strength cements RS, normal and rapid hardening cements N and R, and slowly
hardening cements SL.
Figure 3.6. Compressive strength development of concrete made with various types of cement according to Eq. EC33
Fig 3.6 shows, that the gain in strength in 6 months ranges from 12% for rapid hardening cements to
25% for slowly hardening cements. So, a considerable part of the sustained loading effect is
compensated for by the increase in strength.
Furthermore the bearing capacity as formulated in building codes is generally based on experiments
in laboratories (shear, punching, torsion, capacity of columns). Normally those tests have a duration of
at least 1.5 hours. In Fig. 3.5 it can be seen that in a test with a loading duration of 100 minutes, the
reduction of strength with regard to 2 minutes is already about 15%.
A certain sustained loading effect is therefore already included in the results of tests. It is therefore
concluded that cases in which the sustained loading effect will really influence the bearing capacity of
a structure in practice are seldom and do not justify a general reduction of the design strength with a
sustained loading factor of 0.8. Therefore in clause 3.1.4 it is stated that “the value of Įcc may be
assumed to be 1, unless specified otherwise”.
Such a case can for instance occur when, according to 3.1.2, the concrete strength is determined
substantially after 28 days: in such a case the gain in strength may be marginal so that a value Įcc
smaller than 1 is more appropriate.
For tension similar arguments apply.
The value of the design tensile strength fctd is defined as
fctd = oct fctk 0,05 /¸C [(3.16)EC2] (2.3)
where
Įct is a safety factor, similar to Įcc, the value of which may be found in the National Annex;
fctk 0,05 is the characteristic axial tensile strength of concrete, fractile 5%, which can be deducted from
Table [3.1EC2].
C3.1.7 Stressstrain relations for the design of crosssections
Design stressstrain relations can be derived from the mean stress strain relations. This can
principally be done using the relations given in 3.1.5, but now for the characteristic compressive
strength (fck) values and subsequently reducing the stress ordinate by a factor ¸c = 1.5. This means
that principally not only the stress values but also the Ec values are divided by ¸c. In order to obtain
consistent and sufficiently safe design relations the ultimate strains İcu have also been slightly
reduced in relation to Eq. 8. By choosing the expression
İcu (‰) = 2.6 + 35 [(90 – fck)/100]
4
< 3.5 (9)
for concrete strength classes C55 and higher, all curves end approximately at their top (compare Fig.
3.3). The resulting curves are shown in Fig. 3.7.
Guide to EC2 Section 3
Page 37 Table of contents

Figure 3.7. Basis for design diagrams
In the new version for EC2 two alternative design curves are given: one on the basis of a parabola
rectangle relation (Fig. 3.8), and one on the basis of a bilinear relation (Fig. 3.9)
Figure 3.8. Parabolarectangle design stressstrain relation
Figure 3.9. Design stressstrain curves for a parabolarectangle formation
The parabola rectangle relation (Fig. 3.8) is expressed by
1 1
n
c
c cd
cu
f
(
  c
( o = ÷ ÷

c
(
\ .
¸ ¸
for
2
0
c c
s c s c
c cd
f o = for
2
0
c c u
s c s c
where n, İc2 and İc2u follow from the table below:
Table 3.4. Parameters for the parabolarectangle design stressstrain relation in compression
C20 C35 C50 C55 C60 C70 C80 C90
cc2(‰) 2,0 2,0 2,0 2,2 2,3 2,4 2,5 2,6
ccu2(‰) 3,5 3,5 3,5 3,1 2,9 2,7 2,6 2,6
n 2 2 2 1,75 1,6 1,45 1,4 1,4
Guide to EC2 Section 3
Page 39 Table of contents
on the basis of the parabola – rectangle approach and the bilinear approach.
The figure shows that there is good overall consistency. It could be argued that for the lowest
concrete strength classes (see the figure for C20 in Fig. 3.12), the bilinear approach is slightly less
accurate. This could be improved by reducing the values İc3 for the lower strength classes. This would
however have a considerable disadvantage since then the shape of the design stress strain curves for
all strength classes < C55 would be different which would render
substantial complications for practical design (different shape factors and distance factors, large sets
of design diagrams for various cases, where now only one diagram is sufficient).
Since the loss of accuracy for practical calculations is very small the constant value İc3 = 1.75 %
should be maintained.
Figure 3.12. Comparison of “basic” and “approximate” design curves
In clause 3.1.6 (4) of EC2 as well the possibility is offered to work with a rectangular stress
distribution, see fig. 12. This requires the introduction of a factor Ȝ for the depth of the compression
zone and a factor Ș for the design strength, see Fig. 3.13.
Figure 3.13. Rectangular stress distribution
As a basis for the derivation of ì and Ș the parabolicrectangle stress strain relation is used, see Fig.
3.11. As an example the values ì and Ș are calculated for the strength classes C50 and lower. For
concrete’s in the strength classes C50 the characteristic strains are İc2 = 2.0 ‰
and İc2u = 3.5‰. Now a rectangular stress block is searched for, which gives the same resulting force
at the same location.
Figure 3.14. Derivation of rectangular stress block from the parabolicrectangle stress distribution for
concrete strength class C50
Guide to EC2 Section 3
Page 310 Table of contents
For the parabolicrectangle stress distribution the resulting force is 0.81 xbfcd and the distance of this
force to the top is 0.415x, where x is the height of the compression area.
In order to obtain a rectangular stress block with its resultant at the same position, the depth of the
compression area should be Ȝx = 2*0.415x = 0.83x, so Ȝ = 0.83.
In order to get the same magnitude of the resultant, the maximum stress is defined as Șfcd.
The resultant force for the rectangular stress block is (Ȝx)b(Șfcd). Since this force should be equal to
0.83xbfcd, the value of Ș follows from Ș = 0.81/Ȝ = 0.98
Carrying out this calculation for all concrete strength classes, with the values for İc3 and İc3u taken
from table 3.5 [table 3.1 EC2], the values Ș and ì shown in fig. 3.15 are obtained.
Figure 3.15. Values Ș and Ȝ for the definition of the rectangular stress block
Approximate equations for Ȝ and Ș are:
0,8 ì = per fck 6 50 N/mm
2
( ) 50
0,8
400
ì =
ck
f
per 50 < fck 6 70 N/mm
2
1 q= per fck 6 50 N/mm
2
( )  50
1, 0
200
q =
ck
f
per 50 < fck 6 70 N/mm
2
If the width of the compression zone decreases in the direction of the extreme compression fibre, the
values given in Eq. 16a,b do not hold. This is investigated in Fig. 3.16.
Figure 3.16. Derivation of rectangular stress block from bilinear stress block for a cross section width in the direction of
the extreme compression fibre
Similar type of calculations has been carried out for this case. Fig. 3.17 shows for an Ș as a function
of the concrete strength. Three cases are considered:
 the calculated relations for a rectangular crosssection
 the calculated relations for a triangular crosssection with the top at the extreme compressive strain
 the design equation, derived for the case of the rectangular crosssection.
It is shown that the design equation for Ȝ is safe for both cases. However, the values for Ș for the
triangular case are lower than the design values.
It is sufficient to reduce the values for Ș with 10% in order to cover as well the triangular case.
Guide to EC2 Section 3
Page 311 Table of contents
3.1.8 Flexural tensile strength
3.1.9 Confined concrete
3.2 Reinforcing steel
3.2.1 General
3.2.2 Properties
3.2.3 Strength
3.2.4 Ductility characteristics
3.2.5 Welding
3.2.6 Fatigue
Figure 3.17. Calculated values for and Ȝ for a rectangular and triangular crosssection, in comparison with the design
equation for the rectangular case.
C3.1.8 Flexural tensile strength
The flexural tensile strength is larger than the concentric tensile strength. This is caused by strain
softening at cracking concrete. The size effect of concrete in bending is therefore the same as for
concrete subjected to shear or punching (see f.i. Walraven, 1995), so that the size
factor
200
1 k
d
= + may be expected to apply for the relation between flexural tensile strength and
concentric tensile strength as well. However, in this case the phenomenon is more sensitive to the
effect of drying shrinkage, temperature gradients and imposed deformations.
Therefore the more conservative equation
,
1,6
100
 
= ÷ >

\ .
ctm fl ctm ctm
h
f f f
has been chosen.
C3.1.9 Confined concrete
Having a uniform radial compressive stresses ı2 at the ULS as a result of confinement, the axial
strength fck,c is:
fck,c = fck [1+5,000 (ı2 /fck)] for ı2 0,05 fck [(3.24)EC2] (2.4)
fck,c = fck [1,125 + 2,5 (ı2 /fck)] for ı2 > 0,05 fck [(3.25)EC2] (2.5)
Axial strength increases by 50% if the lateral compressive stresses are 15% of fck, it doubles if
compressive stresses are 35% of fck. Also the strain at failure increase up to 34 times.
Lateral stresses may be achieved by confinement of the compressed member. Index c after fck ,
stands for "confined".
C3.2.2 Properties
Clause [3.2.2(3P)EC2] states that the design rules of Eurocode are valid when steel having
characteristic yield fyk between 400 and 600 N/mm
2
is used.
Guide to EC2 Section 3
Page 312 Table of contents
3.2.7 Design assumptions
3.3 Prestressing steel
3.3.1 General
3.3.2 Properties
3.3.3 Strength
3.3.4 Ductility characteristics
C3.2.7 Design assumptions
Safety factor ¸S for ultimate limit states is 1,15 according to Table [2.1NEC2].
The design stressstrain diagram is shown in Fig. 3.18.
Figure 3.18. Idealised and design stressstrain diagrams for reinforcing steel [Fig. 3.8EC2]
For normal design, either of the following assumptions may be made:
c) an inclined top branch with a strain limit of İud and a maximum stress of
yk
S
f
k
¸
where
t
y
k
f
k =
f
 


\ .
d) a horizontal top branch without the need to check the strain limit.
The recommended value of İud is 0,9 İuk.
The value of
t
y
k
f
f
 


\ .
, given in Annex C for class C steel, is between 1,15 and 1,35.
The value of the elasticity modulus Es may be taken as 200000 Nmm
2
.
C3.3 Prestressing steel
Data from EC2, integrated with those from EN 10138 which is referred to in EC2, are recalled hereafter.
Prestressing steel are geometrically classified as:
wires with plain or indented surface, of diameter between 3,0 and 11,0 mm
twowire strands spun together over a theoretical common axis; nominal diameter of the strand
between 4,5 and 5,6 mm
threewire strands spun together over a theoretical common axis; nominal diameter of the strand
between 5,2 and 7,7 mm
sevenwire strands of which a straight core wire around which are spun six wires in one layer;
nominal diameter of the strand between 6,4 and 18,0 mm
ribbed bars; nominal diameter between 15,0 and 50,0 mm.
Within each type, reinforcing steel is classified according to the following properties:
Strength, denoting the value of tensile strength fp and the value of the 0,1% proof stress (fp0,1k).
Ductility, denoting the value of the ratio of tensile strength to proof strength (fpk /fp0,1k), which should
be at least 1.1, and elongation at maximum load (İuk). Although it's not indicated in EC2, in
accordance with EN10138 İuk should be at least 0,035.
Class, indicating the relaxation behaviour. Three classes are defined in the Eurocode:
• Class 1: wire or strand – ordinary relaxation
• Class 2: wire or strand – low relaxation
• Class 3: hot rolled and processed bars
The design calculations for the losses due to relaxation of the prestressing steel should be based on the
value of ȡ1000, the relaxation loss (in %) at 1000 hours after tensioning and at a mean temperature of 20
°C. The value of ȡ1000 is expressed as a percentage ratio of the initial stress and is obtained for an initial
stress equal to 0,7fp, where fp is the actual tensile strength of the prestressing steel samples. ȡ1000 values
indicated in EC2 for structural design are: 8% for Class 1, 2,5% for Class 2, 4% for Class 3.
Clause [3.3.2(7)EC2] gives the formulae for calculation of relaxation at different t times for the three
abovementioned classes. Annex [DEC2] provides the elements needed for accurate calculations.
Fatigue: prestressing tendons are liable to fatigue. Relevant criteria and methods for verification are
Guide to EC2 Section 3
Page 313 Table of contents
3.3.5 Fatigue
3.3.6 Design assumptions
3.3.7 Prestressing tendons in
sheaths
3.4 Prestressing devices
given at clause [6.8EC2].
On top of strength and ductility values, the Eurocode provides the following design assumptions:
Modulus of elasticity recommended for strands: 195000 N/mm
2
; for wires and bars: 205000
N/mm
2
Design stressstrain diagrams. As represented in Fig. 3.19, taken the safety factor S = 1,15, the
design diagram is made of a rectilinear part up to ordinate fpd from which two ways start: a
rectilinear inclined branch, with a strain limit İud = 0,9 İuk, or 0,02; the other branch is a horizontal
branch without strain limit.
Figure 3.19. Idealised and design stressstrain diagrams for prestressing steel [Fig. 3.10EC2]
LITERATURE
Aitcin, P., “Demystifying Autogenous Shrinkage”, Concrete International, 1999, pp. 5456.
CEBBulletin 228, “High Performance Concrete: Recommended Extensions to Model Code
90”, Report on the CEBFIP Working Group on High Strength Concrete”, July 1995.
ENV 1971 Cement – composition, specifications and conformity criteria – Part 1 Common
cements, 1992
fib, “Structural Concrete: Text Book on Behaviour, Design and Performance, Upgraded
Knowledge of the CEB/FIP Model Code
Han, N., “Time dependant behaviour of high strength concrete”, PhD thesis, Delft University
of Technology, The Netherlands”, 1996.
JCI Technical Committee Report on Autogenous Shrinkage Autogenous Shrinkage of Concrete:
Proceedings of the International Workshop, ed. Tazawa, E&FN Spon, London, 1998,
pp. 363.
Koenders, E., van Breugel, K.,”Second Stichtse Bridge – Concrete with high strength: experimental
research into the creep and relaxation behaviour during the early stage of the hardening
process”, Report 25.5962, Feb. 1996, TU Delft (in Dutch).
Müller, H.S., Küttner, C.H., ”Creep of High Performance Concrete – Characterisation and
Code Type Prediction Model”, 4th International Symposium on Utilisation of Highstrength/
High Performance Concrete, Paris, 1996, pp. 377/386
Müller, H.S., Küttner, C.H., Kvitsel, V., “Creep and shrinkage models for normal and high
performance concrete – a unified code type approach”, Revue Francaise du Genie Civil
(1999)
Neville, A.M., “Properties of concrete”, Pitman Publishing Limited, London 1995,
Guide to EC2 Section 3
Page 38 Table of contents
In Table 3.4 the values for İc2u follow from Eq. (69)
The values for İc2 and n are obtained by curve fitting to the relations shown in Fig. 3.7. An
approximate expression for İc2 is:
İc2(‰) = 2.0 for fck 50 Mpa (12a)
İc2 (‰) = 2.0 + 0.085 (fck – 50)
0,53
for fck > 50 Mpa (12b)
and for n:
n = 2.0 for fck 50 Mpa (13a)
n = 1.4 + 23.4 [(90 – fck)/100]
4
for fck > 50 MPa (13b)
The resulting design curves are shown in Fig. 3.10.
A second possibility is the use of a bilinear design stressstrain relation (Fig. 3.10).
The values İc3 and İc3u are obtained by curve fitting to the relations shown in Fig. 3.7, and are given in
Table 3.5.
Table 3.5. Parameters for the bilinear design stressstrain relation in compression
The values for İc3u are the same as for İc2u and follow therefore from Eq. 9. An approximate
expression for İc3 is
İc3 (‰) = 1.75 for fck < 55 MPa (14a)
İc3 (‰) = 1.75 + 0.11 (fck – 55)0.42 for fck 55 MPa (14b)
The design stressstrain relations derived in this way are shown in Fig. 3.11.
Figure 3.10. Bilinear design stressstrain relation
Fig.3.11. Design stressstrain curves with bilinear formation
Fig. 3.12 shows the comparison for the “basic” design stress – strain relations (derived from the mean
curve by taking the characteristic value and dividing it by ¸c = 1.5) and the “simplified” design curves,
C20 C35 C50 C55 C60 C70 C80 C90
cc3(‰) 1,75 1,75 1,75 1,75 1,90 2,1 2,2 2,3
ccu3(‰) 3,5 3,5 3,5 3,1 2,9 2,7 2,6 2,6
Guide to EC2 Section 3
Page 314 Table of contents
Nilson, A., “High Strength Concrete An Overview of Concrete Research”, Proceedings of
the Conference Utilisation of High Strength Concrete”, Stavanger 1987.
Walraven, J.C., “High Performance Concrete Bridges”, a European Perspective”, Conference
on Developments in Short and Medium Span Bridge Engineering, Calgary, Canada, ’98, Proceedings,
Part I, pp 114.
Walraven, J.C., “Size effects: their nature and their recognition in building codes”, STUDI E
RICERCHE – Vol. 16, 1995, pp. 113134.
Guide to EC2 Section 4
Page 41 Table of contents
SECTION 4 DURABILITY AND
COVER TO REINFORCEMENT
SECTION 4. DURABILITY AND CONCRETE COVER
C4 The rules on design for durability in EC2 are substantially different than in the past. Previously
the concrete cover was prescribed in dependence of the environmental class, but independent of
the concrete quality. In the actual version of EC2 (EN 199211) the cover required depends not
only on the environmental class, but as well on the concrete strength class, the required design
working life and the quality control applied. In the following, background information is given with
regard to those choices. The values for the cover in EN 199211 are a result of increased
understanding in the processes of deterioration, as revealed by the Duracrete studies [1], and
practical experience. In the following theoretical considerations the most important deterioration
processes are explained and parameter studies illustrate the mutual dependencies. Further
background information is found in FIB Bulletin n. 34 “Model code for service life design” [2].
4.1.1 Introduction
In a European research project [1] a probabilistic based durability design procedure of concrete
structures has been developed with the objective to set up a similar concept as in structural design,
where the resistance of the structure is compared to the acting load. Related to corrosion protection
of reinforcement the resistance of a concrete structure is mainly determined by the thickness and
the quality of the concrete cover.
In the European design code for concrete structures EN 199211 the designer has to determine the
nominal concrete cover, which consists of the minimum concrete cover (dependent on the relevant
environmental class) plus an allowance in design for tolerance. According to the first draft of prEN
19921 the allowance in design for tolerance was also dependent on the environmental class – 10
mm for XC0 and XC1 and 15 mm for all other classes. This rule was changed with respect to the
general requirement that values or rules specified in other Eurocodes should not explicitly given
again in EN 199211 but they should only be referred to. Thereafter in the December 99 draft of
prEN 19921 the allowance in design for tolerance was determined with respect to the execution
standard prENV 13670 where the execution tolerance is uniformly defined to 10 mm. This means a
reduction of the nominal cover if the values for the minimum cover are not increased accordingly.
In order to find out which value for the nominal concrete cover is adequate a durability design was
performed, based in the model in the European project. In particular, the reliability index  was
determined and evaluated for different concrete mixes and different nominal concrete covers. The
concrete mixes have been chosen with respect to the environmental classes given in EN 2061.
4.1.2 Description of Deterioration Models and Probabilistic Durability Design
This chapter only gives a short overview on those models, since the deterioration models related to
reinforcement corrosion and the safety concept of durability design have already been thoroughly
described in the literature.
4.1.2.1 Deterioration Models
The corrosion process can be divided into two time periods: the initiation period describes the time
until the reinforcement is depassivated either by carbonation or by penetrating chlorides reaching a
critical chloride content.
After depassivation, corrosion will start if sufficient oxygen and moisture are available. As a result of
corrosion a reduction of the steel cross section, cracking or spalling of the concrete cover will occur.
This time period is described as the propagation period.
In the literature deterioration models [2], by which the processes of the initiation period can be
described, are well established. The process of the propagation period is much more complex and
so far no unanimously accepted models exist.
In the following the timedependent description of the carbonation progress and the time dependent
diffusioncontrolled penetration of chlorides are briefly presented.
4.1.2.1.1 CarbonationInduced Corrosion
The CEB Task Group V model by which the carbonation process in the initiation period can be
predicted is given in equation (4.1):
( )
w
e c Eff,0 0
c
2 k k D ǻC t
x t = t
a t
 

\ .
(4.1)
where:
xc(t) is the carbonation depth at time t
DEff,0 effective diffusion coefficient of dry concrete for carbon dioxide in defined
environment (20°C, 65% rel. humidity)
Guide to EC2 Section 4
Page 42 Table of contents
a the amount of CO2 for complete carbonation [kgCo2/m³]
AC the concentration difference of CO2 at the carbonation front and in the air, which
usually means the carbon dioxide content of the surrounding air c0
ke parameter for micro climatic conditions, describing the mean moisture content of
concrete
kc parameter to describe the curing conditions
w parameter (exponent) for micro climatic conditions at the concrete surface,
describing wetting and drying
t0 reference period,
t law valid (e.g. 1 year) t time
4.1.2.1.2 ChlorideInduced Corrosion
The model for predicting the initiation period in the case of chlorideinduced reinforcement corrosion
is defined by equation (4.2):
( )
( )
n
0
t RCM,0 e crit
t
x t = 2C k D k t
t
 

\ .
(4.2)
ktDRCM,0 = D0 (4.3)
( )
1 crit
crit
SN
C
C = erf 1
C
 

\ .
(4.4)
where:
x(t) depth with a corresponding chloride content (here C(crit)) at time (t)
D0 effective chloride diffusion coefficient under defined compaction, curing and
environmental conditions, measured at time t0 DRCM,0 chloride migration coefficient
under defined compaction, curing and environmental conditions, measured at time t0
Ccrit chloride threshold level
n factor which takes the influence of age on material property into account
kt constant which transforms the measured chloride migration coefficient DRCM,0
into a chloride diffusion coefficient D0
ke constant which considers the influence of environment on D0
erf
1
inverse of the error function
CSN surface chloride level
t time
t0 reference period (28 days)
4.1.2.2 Probabilistic Durability Design
4.1.2.2.1 Safety Concept
The simplest design problems have only one resistance variable and one action variable. They are
generally solved by facing the two variables R and S:
Z = R – S
(4.5)
where Z is the reliability of the structure, R the resistance and S the action: both variables R and S
have their averages and standard deviations, and in this example they are normally distributed. Z is
a variable itself (see Figure 4.1) and is also normally distributed with a mean µz and a standard
deviation oz according to equations (4.6) and (4.7).
µ
Z
= µ
R
÷ µ
S
(4.6)
2 2
z R S
o = o + o
(4.7)
In this simplest case with two variables, the reliability index is the difference between the mean
values of R and S divided by the standard deviation of the variable Z or, alternatively the mean
value of Z divided by the standard deviation of Z (see Equation 4.8, Figure 4.1):
Guide to EC2 Section 4
Page 43 Table of contents
( )
f
p
z
z
  µ
= u ÷ = u ÷

o
\ .
where u(.) stands for a normal distribution and pf for the failure probability.
Figure 4.1 resistance, action, failure probability and reliability index
The numerical problems which are treated in the context of durability design are not as easy to
calculate as this, because there are numerous variables which have to be statistically evaluated and
they are often part of nonlinear functions. Besides these aspects, variables are not always normally
distributed. For this reason, computer programs are used to calculate these kinds of problems.
4.2. Definition of reliability levels
As already mentioned, the corrosion process can be divided into two time periods. For both periods,
limit state conditions can be defined. Depassivation of the reinforcement (= end of the initiation
period) is defined as a serviceability limit state (SLS), because after the end of the initiation period
the corrosion process starts [1]. As the corrosion process in the propagation period can lead to
severe consequences (loss of safety of people and structure), this situation is defined as ultimate
limit state (ULS). Since the calculations performed in this study only concern the initiation period,
the durability design presented here durability design is a SLS assessment.
Design limit states are often defined by means of the reliability index . According to EN1990, the
reliability index for SLS is determined to  = 1,5; in other National Standards the reliability index is
even higher. In the case of durability design a risk oriented grading of the reliability index is
proposed. Because of the fact that only the initiation period is considered, the corrosion process
itself is so far not included in durability design. A possible way of taking account of the different
corrosion risks is to adjust the reliability index to the environmental classes. As a consequence, with
respect to the corrosion process a moderate humid environment (e.g. XD1) or environments with
cyclic wetting and drying (XC4, XD3, XS3) should fulfil higher safety requirements than totally dry or
wet environments (e.g.XC1, XC2, XC3, XS2, XD2) and therefore a higher reliability index is
proposed. For the same reason in chloride environments higher reliability indices should be applied
due to the risk of higher corrosion rates compared to carbonation induced corrosion. A proposal is
given in Table 4.1.
Environmental class Reliability index SLS,50
XC4, XD1, XS1, XS3, XD3 2,0
XC2, XC3, XS2, XD2 1,5
XC1 0,5
Table 4.1: Required reliability index bSLS,50 for a life time of 50 years
4.3 Parametric Study
4.3.1 General
The calculations are based on the models given in 4.2.1.1 and 4.2.2.2. All input values are given as
stochastic variables (mean, standard deviation, type of distribution). The output of the program is
the reliability index  as a function of the lifetime. In this study the reliability index  is considered
for a lifetime of 50 years.
4.3.2 Carbonation
The type of cement has a strong influence on the concrete resistance against carbonation.
Concretes with Portland cement are exposed to carbonation they show a decreasing porosity in the
carbonation zone resulting in higher resistance against ongoing carbonation. In concretes with
blastfurnace cements or with cements with pozzolanic additions (e.g. fly ash), on contrary, the
porosity of the concrete increases after carbonation and the carbonation of the concrete sometimes
progresses faster (as the diffusion of CO2 takes place through the carbonated zone). The influence
Guide to EC2 Section 4
Page 44 Table of contents
of different types of cement is not considered. In the calculations two different concrete mixes
(favourable and unfavourable) were studied.
The w/c ratio of the concrete mixes were chosen according to the requirements in prEN 2061. The
cement content was kept constant (320 kg/m³) for all concrete mixes, because there is only a minor
influence on the carbonation process. The material resistance was determined under laboratory
conditions. The length of the curing period also has a strong influence on carbonation. In the
calculations the curing time was assumed to be 2 days for CEM I and 3 days for CEM III/B. The
climatic conditions were determined according to statistical data from local weather stations. All
calculations were performed for European locations with moderate and hot climate to show the
influence of temperature.
4.3.2.1 Environmental class XC2
In the following, durability design calculations for nominal covers of 30 and 35 mm, for different
climates and for different cements (CEM I and for CEM III/B) are performed. In Figure 4.2 the
results of the calculations are shown.
Figure 4.2: Reliability index versus concrete cover for environmental class XC2 (w/c=0,6,service life 50 years)
a) for CEM I 42,5 R (curing 2 days) b) for CEM III/B 42,5 NW HS NA (curing 3 days)
It can be seen that the reliability index for the calculated values depends on the concrete cover, the
location and the type of cement. Especially in the case of a CEM III/B, the reduction of the concrete
cover by 5 mm results in a decrease of the reliability index of almost 50 %. This means a reduction
of life time of more than 15 years.
It is interesting to see that the choice of a different type of cement has a bigger influence than the
reduction of the concrete cover. However it needs to be taken into account that part of the effect
depending by the type of cement can be counteracted by prolonged curing.
4.3.2.2 Environmental class XC3
For environmental class XC3 the same concrete covers are required as for XC2, but the required
w/c ratio for class XC2 is higher. It can be seen, that this improvement of the concrete quality
(porosity) also has an important effect on the carbonation (compare Figures 4.2 and 4.3).
Apart from this the same trends can be observed. The reduction of the nominal concrete cover from
35 to 30 mm has a big influence on the reliability index, even more pronounced than for concretes
with a w/c ratio according to the requirements for XC2.
Figure 4.3: Reliability index versus concrete cover for environmental class XC3 (w/c=0,55, service life 50 years)
a) for CEM I 42,5 R (curing 2 days) b) for CEM III/B 42,5 NW HS NA (curing 3 days)
4.3.2.3 Environmental class XC4
For environmental class XC4, the w/cvalue is reduced to 0,5 and compared to the environmental
classes XC2 and XC3 the concrete cover is increased by 5 mm. This has a considerable effect on
the reliability index (compare Figures 4.1a, 4.2a and 4.3a). Although the required reliability index
according to Table 4.1 is increased to 2,0 for XC4 all concrete covers are still above the value for
Guide to EC2 Section 4
Page 45 Table of contents
CEM I.
However when using a CEM III/B type of cement, (see Figure 4.3b), the reliability index is reduced
by 50% and is far below the reliability index proposed for this environmental class. An additional
reduction of the concrete cover means a further reduction of the reliability index.
The considerable difference in reliability index as a consequence of using two types of cements
shows that the currently existing requirements for the determination of the concrete cover do not
include all important parameters and are therefore not very exact. For a nominal concrete cover of
40 mm, the model seems to put in evidence that the deviation of the proposed reliability index is still
acceptable, whereas a nominal concrete cover of 35 mm leads to an unacceptable decrease of total
life time.
Figure 4.4  Reliability index versus concrete cover for environmental class XC4(w/c=0,50)
a) for CEM I 42,5 R b) for CEM III/B 42,5 NW HS NA
4.3.3 Chloride penetration
Chlorides are transported in the concrete by the pore water. Processes of diffusion and conveyance
by water transport take place which lead to a certain chloride content in the concrete. The time
before attainment of a critical chloride content depends mainly on the porosity of the concrete,
which may be influenced for example by the w/c ratio and the type of cement. In general, the use of
Portland cements leads to a higher permeability of the concrete for chlorides than the use of
blastfurnace cements or cements with fly ash.
In this study only environmental class XS3 with nominal concrete covers of 50 and 55 mm was
studied. The w/c ratio of the corresponding concrete mix was determined according to EN 2061
(w/c = 0,45) and the cement content was 320 kg/m³. The climatic conditions (relative humidity and
temperature) were determined according to statistical data of local weather stations and the
calculations were performed for two European locations close to the sea in moderate and hot
climate.
The calculations were performed with two different types of cement [1]: the results for a CEM I with
fly ash are shown in Figure 4.4. Alternatively the calculation was performed with CEM I without fly
ash [1]. The reliability index for a lifetime of 50 years was below 0, that means that corrosion
probability is higher than 50%. Examples have shown that the use of Portland cements in tidal
environments (e.g. harbours) has already led to deterioration (cracking) as a result of chloride
induced corrosion after 10 years.
The calculation with Portland cement plus fly ash has shown similar trends as in the case of
carbonation. Figure 4.5 shows that hot climates result in faster chloride ingress than moderate
climates. It can also be seen, that the proposed reliability index is not reached for environmental
class XS4.
Figure 4.5: Reliability index versus concrete cover for environmental class XS3 (w/c=0,45)
a) for CEM I with fly ash
Guide to EC2 Section 4
Page 46 Table of contents
The results of chapters 3.2 and 3.3 have shown that lower values for the nominal concrete cover
result in a significant reduction of the reliability index, leading to a decrease of the safety against
reinforcement corrosion. It can be concluded that the nominal concrete cover should not be
decreased as a consequence of the reduction of allowance in design for tolerance.
4.4 Conclusions
The calculations have shown the dependence of the rate of deterioration on the service life, the
thickness of the concrete cover and the environmental class. It is shown that the provisions as given
in EC2 recognize the most important influencing factors. Further influences, which have not yet
been regarded in detail in EN 199211 are the type of cement and the temperature.
The comparisons show that, theoretically, the prescribed reliability indexes are not always met. It
should be realized, however, that there are many uncertainties in the input values of the
calculations. Further to the influence of the type of cement and the temperature there is the
variation in climatic conditions (wet  dry cycles, local differences due to different orientation with
regard to solar radiation and wind).
The recommendations in the code are the result of theoretical considerations and engineering
experience. In some respects the deterioration models give valuable information. It is shown for
instance that prolonging the service life of a structure from 50 to 100 years requires globally an
increase of the concrete cover between 8 and 12 mm. The advised increase of 10mm is therefore a
good average value, regarding the many unknown factors. Moreover, research is necessary to
close the gap between scientific models and practical observations.
See example n. 4.1, 4.2, 4.3, 4.4.
Reference
[1] DuraCrete – Probabilistic Performance Based Durability Design of Concrete Structures:
Statistical Quantification of the Variables in the Limit State Functions. Report No.: BE 951347,
pp. 6263, 2000.
[2] FIB Bullettin n. 34 Model Code for Service Life Design, June 2006, ISBN 9782883940741
Guide to EC2 Section 5
Page 51 Table of contents
SECTION 5 STRUCTURAL
ANALYSIS
5.1 General
5.2 Geometric imperfections
SECTION 5. STRUCTURAL ANALYSIS
C5.1 General
Structural analysis is the process of determination of the effects of actions (forces, impressed strain)
in terms of tensional states or strain on a geometrically and mechanically defined structure.
The analysis implies a preliminary idealisation of the structure, based on more or less refined
assumptions of behaviour. There are four types of idealisations:
 linear elastic behaviour that assumes, for analysis, uncracked cross sections and perfect
elasticity. The design procedures for linear analysis are given in [5.4EC2];
 linear elastic behaviour with limited redistribution [5.5EC2]. It is a design (not analysis)
procedure based on mixed assumptions, derived from both the linear and nonlinear
analysis.
 plastic behaviour. Its kinematic approach [5.6EC2], assumes at ultimate limit state the
transformation of the structure in a mechanism through the formation of plastic hinges; in its
static approach, the structure is represented by compressed and tensioned elements (strut
and tie model);
 non linear behaviour, that takes into account, for increasing actions, cracking, plastification
of reinforcement steel beyond yielding, and plasticization of compressed concrete. The
design procedures for nonlinear analysis are given in [5.7EC2];
The rules in the EN are technically rather similar to those in the ENV. However, the discontinuities
mentioned above have been removed, and the rules have been coordinated between EC2, EC3
and EC4, see 5.2.5. The EN rules are not repeated here.
C5.2. Geometric imperfection
5.2.1 Symbols
5.2.2 Imperfections and tolerances
The minimum value of the basic inclination is now 1/300 instead of 1/200 and 1/400 (with and
without 2
nd
order effects respectively). A good reason for having one value, independent of the
importance of 2
nd
order effects, is to avoid the accumulation of discontinuities. The present value is
also well correlated with the tolerances given for class 1 in EN 13670, see figure 1. An upper limit
1/200 for the basic value has also been added.
Figure 5.1. Comparison between imperfection and
tolerances. The thick lines represent imperfections, the
black line represents basic imperfection u0 according to
expression 5.1 in EN199211, whereas the grey line
represents the lower limit of the mean value ul for a
structure (m = ).
The thin lines represent the tolerance according to
EN13670; the solid line represents a member in one storey,
the dashed line the total inclination of a structure.
For the total inclination of a column or wall in a structure, the tolerance continues to decrease below
the minimum value 1/300 when the number of storeys exceeds 8. The mean value of the
imperfection will also decrease in a structure where a number of individual members contribute to
the total effect; the lower limit of the mean value (for m = ) is shown in figure 5.1.
The fact that the mean imperfection și is sometimes less than the structural tolerance does not
Guide to EC2 Section 5
Page 52 Table of contents
5.3 Idealisation of the
structure
5.3.1 Structural models for
overall analysis
mean that the imperfection is on the “unsafe side”. The tolerance has to be checked for individual
columns and walls, whereas oim represents the average inclination of all vertical members
contributing to a certain effect.
5.2.3 Equivalent eccentricity or horizontal force
For isolated members, like in the ENV, the inclination can be transformed to either an equivalent
eccentricity (or initial deflection) ei or a horizontal force Hi. This is important for e.g. a pinended
column, where an inclination has no effect on the column itself. The eccentricity can then
represent either an uncertainty in the position of the axial load, or an initial deflection (outof
straightness). The equivalent eccentricity is linked to the effective length, and the horizontal force
should give the same bending moment as the eccentricity (see figure 5.1 in EN chapter 5.2):
Equivalent eccentricity: ei = șiā l0/2 (l0 is the effective length)
Cantilever: ei = șiā l Mi = Nā ei = Nā șiā l = Hiā l ĺ Hi =șiā N
Pinended: ei = șiā l/2 Mi = Nā ei = Nā șiā l/2 = Hiā l/4 ĺ Hi =2șiā N
5.2.4 Dealing with first order effects
Imperfections can be treated as first order effects, or be added as separate safety elements without
any physical meaning, “outside” the second order analysis. With a method like the “curvature
method” (“model column” method in ENV), which gives a fixed second order moment independent
of the first order moment, there is no difference between the two approaches. In other methods,
however, the second order effects depend on the first order effects, and then it does make a
difference whether imperfections are treated as first order effects, or added separately.
In 4.3.5.4 P(1) in the ENV, the imperfection is associated with “uncertainties in the prediction of
second order effects”, which indicates that it is not regarded as a first order effect. The definition of
the first order eccentricity in 4.3.5.6.2 further underlines this. On the other hand, formulations in
2.5.1.3 describe the imperfection rather as a first order effect. Thus, the ENV is ambiguous and
unclear in this respect.
In the EN it is stated once and for all that imperfections are to be treated as first order effects; see
the definition in 5.8.1. This corresponds to a physical interpretation of the imperfection as a
deviation in the form of an inclination, an eccentricity or an initial deflection. This is logical, since
there is a link between imperfections and tolerances. It is essential to have a clear definition in this
respect for the overall analysis of structures, but also for isolated members, when other methods
than the curvature method are used; see 5.8.6 and 5.8.7.
C5.3 Structural models
C5.3.1 Classification of structural elements
For buildings, as a convention, the following provisions apply:
5. a beam is a linear element, for which the span is not less than 3 times the overall section
depth. Otherwise it should be considered as a deep beam.
6. a slab is a bidimensional member for which the minimum panel dimension is not less than 5
times the overall slab thickness. Moreover: a slab subjected to dominantly uniformly
distributed loads may be considered to be oneway spanning if either (Fig. 4.2):
 it possesses two free (unsupported) and sensibly parallel edges, or
 it is the central part of a sensibly rectangular slab supported on four edges with a ratio
of the longer to shorter span greater than 2.
7. ribbed or waffle slabs need not be treated as discrete elements for the purposes of analysis,
provided that the flange or structural topping and transverse ribs have sufficient torsional
stiffness. This may be assumed provided that (Fig. 4.2):
 the rib spacing does not exceed 1500 mm
 the depth of the rib below the flange does not exceed 4 times its width
 the depth of the flange is at least 1/10 of the clear distance between ribs or 50 mm,
whichever is the greater
 transverse ribs are provided at a clear spacing not exceeding 10 times the overall
depth of the slab.
Guide to EC2 Section 5
Page 53 Table of contents
5.3.2 Geometric data
5.3.2.1 Effective width of
flanges (all limit states)
Figure. 5.2.Geometric parameters for slabs
The minimum flange thickness of 50 mm may be reduced to 40 mm where permanent blocks are
incorporated between the ribs. This exception applies for slabs with clay blocks only. It does not
apply for expanded polystyrene blocks.
An exception to this rule is given at [10.9.3(11)EC2] in relation to prefabricated slabs without
topping, which may be analysed as solid slabs provided that the in situ transverse ribs are provided
with continuous reinforcement through the precast longitudinal ribs and at a spacing according to
Table [10.1)  EC2].
A column is a member for which the section depth does not exceed 4 times its width and the height
is at least 3 times the section depth. Otherwise it should be considered as a wall.
C5.3.2 Geometric data
C5.3.2.1 Effective width of flanges of T beams (valid for all limit states)
If a T beam with a relatively wide flange is subjected to bending moment, the width of flange that
effectively works with the rib in absorbing the compressive force (effective width) should be
assessed.
An exact calculation shows that the actual distribution of compressive stresses has a higher
concentration in the part of flange which is close to the rib, and a progressive reduction in the
further parts. This implies that the conservation of plane sections is not respected and that the
neutral axis is not rectilinear, but is higher on both sides of the rib.
In order to simplify calculations, the actual distribution of stresses is usually replaced by a
conventional block, extended to the effective width. This allows the application of the usual design
rules, and in particular the assumption that plane sections remain plane.
Effective width is defined at [5.3.2.1EC2] as a function of the cross section geometry (b, distance
between adjacent ribs; bw, width of ribs) and of the distance lo between points of zero moment. Note
that the flange depth is not relevant, even if it is expressly cited in (1)P, and that the distance
between points of zero moment depends, for continuous beams, on the type of loading Fig. 5.2EC2
is an example of a continuous beam (subjected to a uniform load distribution) where the lo distance
for spans and for parts on supports is identified.
Hence different sections have different effective width. Point (4) makes clear that a constant width
may be assumed over the whole span. The value applicable to the span section should be adopted.
Figure 5.3. Definition of lo for calculation of the effective flange width [Fig. 5.2 – EC2].
Figure 5.4. Parameters to determine effective flange width [fig. 5.3 – EC2]
s s 1500 mm
hf > sn/10 or 50 mm
hw s 4 bm
st s 10 h0
Guide to EC2 Section 5
Page 54 Table of contents
5.3.2.2 Effective span of
beams and slabs in buildings
5.4 Linear elastic analysis
5.5 Linear analysis with
limited redistribution
C5.3.2.2 Effective span of beams and slabs in buildings
This paragraph defines the effective span length mainly for member analysis, taking into account
the different types of support. Two important points must be noted:
÷ Where a beam or slab is monolithic with its supports, the critical design moment at the
support may be taken as that at the face of the support. That moment should not be less
than 65% that of the full fixed end moment.
÷ The design moment and reaction transferred to the supporting element should be taken as
the greater of the elastic or redistributed values.
÷ Regardless of the method of analysis used, where a beam or slab is continuous over a
support which may be considered to provide no restraint to rotation, the design support
moment, calculated on the basis of a span equal to the centretocentre distance between
supports, may be reduced by:
ǻMEd = FEd,sup t / 8
where FEd,sup is the design support reaction
t is the breadth of the support.
This formula derives by assuming an uniform distribution of the design support reaction Fed,sup over
the breadth of the support
Ed,sup Ed,sup
Ed
F F t
t t
ǻM = =
t 2 4 8
 

\ .
C5.4 Linear elastic analysis
For the determination of load effects, linear analysis may be used assuming:
uncracked cross sections,
linear stressstrain relationships and
mean value of the modulus of elasticity.
With these assumptions, stresses are proportional to loads and therefore the superposition principle
applies. For thermal deformation, settlement and shrinkage effects at the ultimate limit state (ULS),
a reduced stiffness corresponding to the cracked sections, neglecting tension stiffening but
including the effects of creep, may be assumed in accordance with [5.4(3)EC2]. For the
serviceability limit state (SLS) a gradual evolution of cracking should be considered.
Comments.
In the previous explanation the expression For the determination of load effects’ was used,
whereas [5.4(3)EC2] admits “For the determination of action effects”, therefore of all actions,
including thermal deformation, settlement and shrinkage for which [5.4(3)EC2] admits different
assumptions, without which the effects of impressed deformations would be devastating and
quantitatively incorrect.
the fact that no limits were set to (xu/d) for the application of the linear analysis method at the
ultimate limit states, does not mean that any value of (xu/d) may be used in design: it's
opportune to observe a limit consistent with the method of linear elastic analysis with limited
redistribution, for which xu/d 0,45. It must be remembered that increasing values of xu/d the
model uncertainty also increases and higher safety factors should be assumed for precaution.
C5.5 Linear elastic analysis with limited redistribution [5.5  EC2]
At ultimate limit state plastic rotations occur at the most stressed sections. These rotations transfer
to other zones the effect of further load increase, thus allowing to take, for the design of
reinforcement, a reduced bending moment įM, smaller than the moment M resulting from elastic
linear design, provided that in the other parts of the structure the corresponding variations of load
effects (viz. shear), necessary to ensure equilibrium, are considered.
Despite being named “Linear analysis with limited redistribution”, this is a design method.
In clause [5.5(4)EC2], in relation with continuous beams and slabs with ratio between adjacent
spans in the range [0,5 – 2] expressions are given for the redistribution factor į in function of the
concrete class, the type of steel and the xu/d ratio after redistribution. For instance, for concrete up
to C50 and reinforcing steel of type B and C, respectively of average and high ductility, the
expression is:
į
>
0,44 + 1,25 (xu/d) ; į
>
0,70 [5.10a – EC2] (4.8)
where:
o is the ratio of the redistributed moment to the elastic bending moment
xu is the depth of the neutral axis at the ultimate limit state after redistribution
d is the effective depth of the section
Guide to EC2 Section 5
Page 55 Table of contents
5.6 Plastic analysis
5.7 Nonlinear analysis
The limits of this formula are:
į = 0,70 per (xu/d) = 0,208 and
į = 1 per (xu/d) = 0,45
It must be considered that a redistribution carried out in observance of the ductility rules only
ensures equilibrium at the ultimate limit state. Specific verifications are needed for the serviceability
limit states. Very high redistributions, which may be of advantage at the ultimate limit states, very
often must be lowered in order to meet the requirements of serviceability limit states. For the
design of columns the elastic moments from frame action should be used without any redistribution.
C5.6 Plastic analysis
Plastic analysis should be based either on the lower bound (static) method or on the upper bound
(kinematic) method for the check at ULS only.
5.6.1 Static method
It is based on the static theorem of the theory of plasticity, which states: “whichever load Q, to which
a statically admissible tension field corresponds, is lower or equal to the ultimate load Qu". The
expression “statically admissible” indicates a field that meets both the conditions of equilibrium and
the boundary condition without exceeding the plastic resistance.
An important application of this method is the strutandtie scheme [5.6.4 – EC2]. Other applications
are the management of shear by the method of varying ș and the analysis of slabs by the
equivalent frame analysis method [Annex I – EC2].
5.6.2 Kinematic method
In this method, the structure at ultimate limit states becomes a mechanism of rigid elements
connected by yield hinges. The method is based on the kinematic theorem, which states: “every
load Q, to which corresponds a kinematically admissible mechanism of collapse, is higher or equal
to the ultimate load Qu".
The method is applied for continuous beams, frames and slabs (in this last case with the theory of yield
lines.
For beams, clause [5.6.2)EC2] states that the formation of plastic hinges is guaranteed provided
that the following are fulfilled:
i) the area of tensile reinforcement is limited such that, at any section
(xu /d) s 0,25 for concrete strength classes s C50/60
(xu /d) s 0,15 for concrete strength classes > C55/67
ii) reinforcing steel is either Class B or C
iii) the ratio of the moments at intermediate supports to the moments in the span shall be
between 0,5 and 2.
If not all the conditions above are fulfilled, the rotation capacity must be verified, by checking the
required rotations against those allowed in accordance with [Fig. 5.6NEC2].
It should be remembered that the plastic analysis methods shall only be used for checking ultimate
limit states. Serviceability limit states requirements should be checked by specific verifications.
C5.7 Nonlinear analysis
Nonlinear analysis is a procedure for calculation of action effects, based on idealisations of the
nonlinear behaviour of materials [nonlinear constitutive laws: for concrete cf. Eurocode 2, 3.1.5(1)
expression (3.14) and Fig. 3.2; for steel 3.2.7(1) Fig. 3.8], of the elements and of the structure
(cracking, second order effects), suitable for the nature of the structure and for the ultimate limit
state under consideration.
It requires that the section geometry and reinforcement are defined, because it is a process of
analysis. Resulting stresses are not proportional to the applied actions.
The process is developed by computeraided calculations, by verifying equilibrium and compatibility
at every load increase. Compatibility conditions are normally expressed by assigning to each
section its moment – curvature law, and integrating the curvatures along the axis of the elements.
Inelastic rotations are generally concentrated in the critical sections. Deformations due to shear are
generally neglected, those in relation with axial load are taken into account only in case have
significant influence on the solution. As the superposition principle does not apply because of the
nonlinearity, the calculations must be developed for each load condition: for each one it is
conventionally assumed that the ultimate limit state is reached through a single proportional
increase of the applied load.
In the case of elements mainly subjected to bending, trilinear idealizations of the moment / rotations
Guide to EC2 Section 5
Page 56 Table of contents
5.8 Analysis of second order
effects with axial load
5.8.1 Definitions
law of each critical section can be adopted as in Fig. 4.6, representing the three following states:
first state (elastic and linear): characterized by EI rigidity of the entirely reacting
sections; it ends when the tensional strength of concrete is reached (cracking moment)
second state (cracked): from the cracking moment to the moment corresponding to
steel yielding, moment increases are related to the curvature increases on the basis of
rigidity
Es As z(dx), where As is the cross section of the tensioned reinforcement, z the lever arm, x
the depth of the neutral axis. The rigidity can be increased by taking into account the
contribution of concrete in tension between cracks (“tension stiffening”), but with caution in
case of load cycles .
third state (plastic): a third linear line can be idealized from the steel yielding clause
to the point of failure moment. The line corresponds to a șpl plastic rotation at the critical
section, with a value that can be deducted from the diagram in Fig 5.6N of Eurocode 2 in
function of the relative depth of the neutral axis. Following the evolution of response to
actions, it is possible to verify the conditions for the serviceability limit states and for the
ultimate limit state.
Figure 5.5. Trilinear Moment – Rotation relation for element mainly subjected to bending
C5.8 Second order effects with axial load
C5.8.1. Definitions
Definitions specific to chapter 5.8 are listed in 5.8.1. Some comments are given below.
Braced – bracing
The distinction braced – bracing is simple: units or systems that are assumed to contribute to the
stabilization of the structure are bracing elements, the others are braced. Bracing units/systems
should be designed so that they, all together, have the necessary stiffness and resistance to
develop stabilization forces. The braced ones, by definition, do not need to resist such forces.
Buckling
The word buckling has been reserved for the “pure”, hypothetical buckling of an initially straight
member or structure, without load eccentricities or transverse loading. It is pointed out in a note that
pure buckling is not a relevant limit state in real structures, due to the presence of imperfections,
eccentricities and/or transverse loads. This is also a reason why the word “buckling” is avoided in
the title of 5.8. In the text, buckling is mentioned only when a nominal buckling load is used as a
parameter in certain calculation methods.
First order effects
First order effects are defined to include the effect of imperfections, interpreted as physical
deviations in the form of inclinations or eccentricities. The ENV is ambiguous in this respect; see
also clause C5.2.
Nominal second order moment
The nominal second order moment is used in certain simplified methods, to obtain a total moment
used for design of cross sections to their ultimate moment resistance. It can be defined as the
difference between the ultimate moment resistance and the first order moment, see 6.3. If the
ultimate load is governed by instability before reaching the cross section resistance, then the
nominal second order moment is greater than the true one; this is the reason for using the word
“nominal”.
Sway – nonsway and global second order effects
The terms sway – nonsway have been omitted in the final draft, after many comments for or
against. The words in themselves are misleading, since all structures are more or less “sway”; a
Guide to EC2 Section 5
Page 57 Table of contents
5.8.2 General
5.8.3 Simplified criteria for
second order effects
5.8.3.1 Slenderness criterion
for isolated members
structure that would be classified as “sway” could be just as stiff as one classified as “nonsway”.
These terms are now replaced by unbraced – braced.
In the ENV the concept of sway – nonsway was linked to the criterion for neglecting global second
order effects in structures. The classification of structures from this point of view remains in the EN,
but without using the “sway”  “nonsway” terminology.
A stiffness criterion like the one in ENVA.3.2 was avoided in earlier drafts of the EN, since it was
considered as too crude, and in some cases misleading. However, during the conversion process
there were many requests to include some simple criterion for evaluating the significance of global
second order effects, without the need for calculating them. This led to the present rules in 5.8.3.3
and Informative Annex H, which are more general than those in ENVA.3.2.
The details are given in clause 3.3.
C5.8.2. Basic criteria for neglecting second order effects
Two basic criteria for ignoring second order effects have been discussed during the conversion
process, namely:
1) 10 % increase of the corresponding first order effect,
2) 10 % reduction of the load capacity, assuming a constant eccentricity of the axial force.
The first criterion is the one stated in 5.8.2 (6), and in the ENV, 4.3.5.1 (5). The second one has
been claimed by some to be the “true”, hidden criterion behind the ENVrules.
Figure 5.6 illustrates the consequences of these two criteria in an interaction diagram for axial force
and bending moment. Their effects on the slenderness limit are discussed in chapter 3.
In a column or a structure it is the bending moment that is influenced by second order effects. The
axial force is governed by vertical loads, and is not significantly affected by second order effects.
Most design methods are based on calculating a bending moment, including a second order
moment if it is significant. From this point of view, criterion 1 is the most logical and natural one.
The basic criterion is further discussed in chapter 3 in connection with slenderness limits.
Figure 5.6. Two different ways of defining the basic 10%criterion for ignoring second order effects, see text above.
(The interaction diagram was calculated for rectangular cross section 400 x 600 mm, concrete C35, Ȧ = 0,1 (total
mechanical reinforcement ratio), edge distance of reinforcement 60 mm.)
C5.8.3. Simplified criteria for ignoring 2nd order effects
C5.8.3.1 Slenderness limit for isolated members
5.8.3.1.1 General
The load bearing capacity of a member in compression for low slenderness ratios is illustrated in
figure 31 by means of interaction curves, calculated according to the general method in 5.8.6.
(See chapter 6 in this report for more details about interaction curves and the general method.)
Guide to EC2 Section 5
Page 58 Table of contents
Figure 5.7. Interaction curves for low slenderness ratios, calculated for a pinended column with cross section as for
figure 21 and subjected to a constant first order moment.
Effective creep ratio ¢ef = 0. Equal first order end moments M01 = M02 = M0.
By combining figures 5.7 and 5.6, one can find the slenderness ratios for which the basic 10%
criterion is fulfilled; see figure 5.8.
Figure 5.8 Effect of normal force (here n = N/fcdAc) on slenderness limit, depending on the interpretation of the 10 %
criterion:
1. 10 % increase of bending moments for a given normal force
2. 10 % reduction of load capacity for a constant eccentricity
(e = 0,3, ¢ef = 0, M01/M02 = 1)
Depending on which of the two basic criteria is chosen, see chapter 2, increasing the axial force will
either decrease (1) or increase (2) the slenderness limit as shown in figure 32. Criterion 1 will be
more severe for high axial loads, when there is little room for bending moments.
Criterion 2, on the other hand, will allow very high slenderness ratios for high axial loads.
In earlier drafts, including the October 2001 “final draft”, a slenderness limit independent of the axial
force was chosen as a compromise between the two basic criteria; for e = 0,1 the criterion was
then identical to expression (4.62) in the ENV.
In a comment to the “final draft” it was pointed out that a limit independent of the axial force could
be much on the unsafe side in certain cases (see next chapter). Therefore, a new model was
developed. The following is quoted from 5.8.3.1:
(1) As an alternative to 5.8.2 (6), second order effects may be ignored if the slenderness ì is
below a certain value ìlim. The following may be used:
ìlim = 20ā Aā Bā C (5.13)
where:
ìslenderness ratio as defined in 5.8.3.2
A = 1 / (1+0,2¢ef)
B = (1+ 2e) / n
Guide to EC2 Section 5
Page 59 Table of contents
C = 1,7  rm
¢ef effective creep ratio; see 5.8.4;
if ¢ef is not known, A = 0,7 may be used
e = Asfyd / (Acfcd); mechanical reinforcement ratio; if e is not known, B = 1,2 / n may be used
As total area of longitudinal reinforcement
n = NEd / (Acfcd); relative normal force
rm = M01/M02; moment ratio
M01, M02 first order end moments,
02 01
M M >
(2) If the end moments M01 and M02 give tension on the same side, rm should be taken positive
(i.e. C < 1,7), otherwise negative (i.e. C > 1,7).
In the following cases, rm should be taken as 1,0 (i.e. C = 0,7):
 for braced members with first order moments only or predominantly due to imperfections or
transverse loading
 for unbraced members in general
The background to the new criteria is presented in the following.
5.8.3.1.2 History of the slenderness limit in prEN 1992
At an early stage of the conversion process (spring 1999), a slenderness limit was proposed in
which the effective creep ratio ¢ef and the relative normal force n were included as parameters.
The reinforcement ratio was not included then, since it was considered impractical.
Most people also considered it impractical and unnecessary to include creep; there was a
widespread opinion that creep would have little effect at these low slenderness ratios.
There was also a lot of discussion about the interpretation of the basic 10% criterion for ignoring
second order effects, with the two main alternatives (see chapter 2):
1. 10 % increase of bending moments due to second order effects
2. 10 % reduction of the load capacity for a given eccentricity
It was then demonstrated that the effect of the normal force on the slenderness limit was different
depending on which alternative was used, see figure 5.8.
However, there was no agreement as to which alternative to base the slenderness limit on, and
therefore the ENV criterion (4.62), independent of the normal force, was used in draft 1.
An addition was made in draft 2, allowing the constant 25 to be increased to 35 if the reinforcement
ratio e is at least 0,5. In the “final” draft October 2001 an interpolation was introduced to avoid
discontinuity (expression (5.13)).
In November 2001, shortly after the draft had been distributed, comments and examples were
presented by Prof. J. Hellesland, showing that (5.13) might be extremely unsafe in some cases, e.g.
a column bent in double curvature (end moments of different directions, figure 5.9), combined with a
high effect of creep and a moderate or high normal force.
Comments on earlier versions of prEN 199211, together with a general treatment of the
slenderness limit, are given in [8].
Figure 5.9. Column bent in double curvature
5.8.3.1.3 Background to the new proposal
After receiving the abovementioned comments and examples from Prof Hellesland, a systematic
investigation of the slenderness limit was made, with focus on the effects of reinforcement, normal
force, creep and moment ratio (different end moments).
Figure 5.10 shows examples of the effect of a rather moderate effective creep ratio, ¢ef =1.
Guide to EC2 Section 5
Page 510 Table of contents
Curves according to both basic 10% criteria are shown. Table 5.1 shows some values taken from
the figures.
Figure 5.10. Slenderness limit as a function of the relative normal force and the moment distribution. Concrete C40, e
= 0,3. (Solid lines = 10%criterion alt. 1, dashed = alt. 2.)
Table 5.1. Values of slenderness limit taken from figure 5.10, alternative 1 (e = 0,3.
For n = 1,0, the average reduction due to creep is considerable, having in mind that ¢ef = 1
represents a rather moderate effect of creep. With a higher value of ¢ef the reduction is more
severe (most of the comparisons were made with ¢ef = 0 and 2 respectively, see Appendix 1).
The effect of e is considerable, as can be seen from both figure 5.11 and table 5.2. Without e as a
parameter, the slenderness limit would have to be either very conservative for high to moderate
values of e, or on the unsafe side for low values. However, since the reinforcement is normally not
known when the slenderness criterion is checked, a default value based on a low value of Ȧ has
also been given. This can be used for a conservative estimation, or as a starting value in an
iterative process.
Figure 5.11. Effect of reinforcement ratio on slenderness limit (only criterion 1 is shown here).
Table 5.2. Values of slenderness limit taken from figure 5.11. ( ¢ef = 2)
There have been national comments proposing to include the effect of n in the slenderness limit.
These proposals were rejected until and including the October 2001 draft, referring to figure 5.8 and
the disagreement concerning the 10%criterion. However, as can be seen from figures 5.10 and
5.11, with different end moments there is a strong reduction of the slenderness limit with increasing
normal force, and that is true for both 10%criteria. The only exception is when criterion 2 is applied
to columns with equal end moments, see figure 5.10.
n M01/M02
ìlim Average increase from
e = 0,1 to e = 0,5 % e = 0,1 e = 0,5
0,4
1,0 17 27
16 0 47 70
0,9 74 107
0,8
1,0 11 20
37 0 27 43
0,9 41 66
n M01/M02
ìlim Average reduction
due to creep, %
¢ef = 0 ¢ef = 1
0,5
1,0 30 25
16 0 70 60
0,9 110 90
1,0
1,0 20 10
37 0 45 30
0,9 70 50
Guide to EC2 Section 5
Page 511 Table of contents
5.8.3.1.4 Comparisons with previous model for slenderness limit
In tables 5.3 and 5.4, values from tables 5.1 and 5.2 are compared with values according to the
slenderness criterion in the October 2001 draft (expression 5.13):
ìlim = 25ā (e + 0,9)ā (2  M01/M02)
In both cases the 10 % criterion is according to alternative 1.
Table 5.3. Slenderness limit in draft Oct. 2001 compared to values from table 5.1 (e = 0,3)
For n = 0,5 and ¢ef = 0 the values according to draft October 2001 are reasonable (table 5.3).
For n = 1,0, however, they overestimate the slenderness limit, particularly for ¢ef = 1 and most
particularly for M01/M02 = 1, where it gives a 3 times too high value.
Table 5.4. Slenderness limit in draft Oct. 2001 compared to values from table 5.2 ( ¢ef = 2).
Draft October 2001 gives a fairly correct influence for e, but it severely overestimates the
slenderness limit for a high normal force combined with a high creep effect, see table 5.4 (here the
values are based on ¢ef = 2). The omission of the effects of both normal force and creep are the
main disadvantages of this model.
5.8.3.1.5 Comparisons with new model
In tables 5.5 and 5.6 the new model (see p. 4) is compared to the same data as in tables 5.1and 5.2
respectively.
Table 5.5. Slenderness limit according to new model compared with table 31 (e = 0,3).
Table 5.6. Slenderness limit according to new model compared with table 5.2 ( ¢ef = 2)
n M01/M02
ìlim Average increase from
e = 0,1 to e = 0,5 % e = 0,1 e = 0,5
0,4
1,0 17 27
16 0 47 70
0,9 74 107
0,8
1,0 11 20
37 0 27 43
0,9 41 66
n M01/M02
ìlim
10% criterion (alt.1)
Draft Oct. 2001
¢ef = 0 ¢ef = 1
0,4
1,0 30 25 30
0 70 60 60
0,9 110 90 87
0,8
1,0 20 10 30
0 45 30 60
0,9 70 50 87
Guide to EC2 Section 5
Page 512 Table of contents
5.8.3.2 Slenderness and
effective length of isolated
members
5.8.3.1.6 Discussion
On the whole, the new model gives good agreement with the 10%criterion (alt. 1), and the main
parameters are taken well into account. There is a slight overestimation of the slenderness limit for
n = 1,0 and ¢ef = 1, table 5.5, also for e = 0,1 and n = 0,8, table 5.6. However, the overestimations
are small compared to the old model (see 3.1.4) and in both cases the values are conservative
compared to 10%criterion alt. 2.
A complete verification of the new model is given in Appendix 1. A more sophisticated model could
of course give even better agreement, e.g. by also including the concrete grade, but the present
model is considered to be good enough. Concrete grade is indirectly taken into account in the n
value.
The importance of considering creep in the slenderness limit is further substantiated in 4.3.3.
C5.8.3.2 Effective length
New expressions (5.15) and (5.16) for the effective length of isolated members in frames, were
introduced in the second draft. Derived to give accurate estimation, based on the definition of
effective length in 5.8.1, they replace figure 4.27 in the ENV as well as expressions (5.22) and
(5.23) in draft 1 of the prEN, December 1999.
The expressions in draft 1 were taken from UK proposals, included in comments on the ENV and on
earlier EN drafts. It was found that they are very conservative in some cases, giving up to 40 %
overestimation of the effective length for braced members and on the unsafe side in other cases,
giving up to 20 % underestimation of the effective length for unbraced members.
It has been claimed that the conservativeness was deliberate, in order to cover certain unfavourable
nonlinear effects. However, the effective length is by definition based on linear behaviour, and the
present models are aimed at giving an accurate estimation according to this, without including some
hidden allowance for possible unfavourable effects. Such effects are instead explicitly addressed in
5.8.3.2 (5) and in 5.8.7.2 (4). The new expressions also avoid unsafe estimations, as in the case of
unbraced members with the previous expressions.
Figure 5.12a and b show comparison between an accurate numerical calculation of the effective
length and estimations according to draft 1 (a) and final draft (b) respectively.
Figure 5.12a. Effective length according to accurate and simplified calculations, draft 1
Figure 5.12b. Effective length according to accurate and simplified calculations, final version
The present kfactors are defined differently compared to the corresponding factors in draft 1 and
ENV, and are called k1 and k2 to avoid confusion with the previous factors ka and kb. The present k
factors express the relative flexibility of the restraint according to the definitions in figures 5.13 and
5.14. They are applicable to different types of flexible moment restraint, such as beams with
different boundary conditions, flexible foundations etc.
Guide to EC2 Section 5
Page 513 Table of contents
Figure 5.13. Flexibility of restraint, example of definition.
Figure 5.14. Comparison between different definitions of kfactors; examples.
5.8.3.2 (4) addresses the question whether an adjacent column (in a storey above or below) in a
node should be considered as using the same restraint as the column considered, or as contributing
to the restraint. This will depend on the magnitude of the axial force in the adjacent column. If both
columns connected to the node reach their respective buckling load at the same time (under
proportional increase of loads), they will both have to share the restraint provided by other
connected members (beams), and k should then be defined as
a c
a c
EI EI
k = +
M l l
( u
(
¸ ¸
(31)
Here subscripts a and c refer respectively to the adjacent column and to the one considered see
Figure 5.15.
In the opposite case, when the adjacent column has a relatively low axial load, it can be included
among the members which resist the moment M, i.e. it will contribute to the restraint.
A reasonable model for the transition between the two limiting cases gives by the following:
( )
a c
1 2 a a c
EI EI
k = +
M +M +...+ 1 M l l
  u
o

o
\ .
(32)
where M1, M2… restraining moments in members 1, 2…, see Figure 5.15
Ma restraining moment in the adjacent column, see Figure 5.15, calculated without
taking into account the axial force Na
o = Na/NBa
Na axial force on the adjacent column
NBa buckling load of the adjacent column (can be estimated approximately, e.g.
taking into account only the horizontal members adjacent to its nodes)
Guide to EC2 Section 5
Page 514 Table of contents
5.8.3.3 Global second order
effects in buildings
Figure 5.15. Illustration of node with adjacent members.
C5.8.3.3 Global second order effects in structures
5.8.3.3.1 Background
a) Model in ENV, A.3.2
According to the ENV, global second order effects may be ignored, and the structure may be
considered as “nonsway”, if
V cm c
L = F E I s o (33)
Where L total height of building (htot in the ENV)
FV total vertical load (FV in the ENV)
EcmIc sum of bending stiffnesses of bracing members
o 0,2 + 0,1ns 0,6 (in the ENV, no particular symbol is used for this parameter)
ns number of storeys (n in the ENV)
This criterion is valid only on certain conditions (not stated in the ENV):
1. No significant rotation at the base (rigid restraint / stiff foundation)
2. No significant global shear deformations (e.g. no significant openings in shear walls)
These conditions are not fulfilled for e.g. a bracing system including frames, shear walls with large
openings and/or flexible foundations.
The criterion also explicitly requires that bracing members are uncracked. In practice, bracing
members are often more or less cracked in ULS, due to high lateral and low vertical loadings (most
of the vertical load is often carried by the braced members).
For the above reasons, ENV A.3.2 has a very limited field of application. Since the limitations of the
applicability are not stated, and no information is given for the cracked stage, there is a also risk
that it is used outside the scope, giving unsafe results.
b) New proposals
Due to the above shortcomings, the ENV criterion was not included in earlier drafts of the EN. After
many requests to include something similar, two alternative proposals were presented to CEN
TC250/SC2 (Berlin, May 2000), a “miniversion” and a “full version”:
• “Miniversion”: same scope as in ENVA.3.2, and a criterion given in a similar closed form.
However, the conditions and restrictions are clearly stated, and the criterion is improved to be less
conservative and to take into account cracking in a simple way.
• “Full version”: formulated in a more general and transparent way. Detailed information is given
only for regular cases, but the formulation opens for general cases. A simple extension is given to
cover the effect of global shear deformations.
In the final draft a somewhat extended “miniversion” is given in the main text, and further extension
to the “full version” are given in Annex H.
5.8.3.3.2 No significant shear deformations, rigid moment restraint
The basic criterion “second order effects 10% of first order effects” gives, together with the
simplified magnification factor for bending moment in 5.8.7.3 (3):
0Ed
Ed 0Ed
V,Ed V,BB
M
M 1,1 M
1 F F
~ s (34)
This gives the following criterion for the vertical load, cf. expression (H.1):
FV,Ed FV,BBā 0,1 / 1,1 § 0,1 FV,BB (35)
Here FV,Ed total vertical load
Guide to EC2 Section 5
Page 515 Table of contents
FV,BB nominal buckling load for global bending (no shear deformations)
M0Ed first order moment
MEd design moment
The global buckling load for bending can be written
2
V,BB 0
EI
F =
L
ç
_
(36)
where ç
0
coefficient depending on number of storeys, distribution of vertical load etc.
EEI total bending stiffness of bracing members; to account for cracking in a simplified way
EEI is based on 0,4āEcdIc; and for uncracked section 0,8 may be used instead of 0,4
L total height
Note FV,BB is a nominal buckling load, calculated for a secant stiffness representing the relevant ULS
conditions (including lateral loading). It is not a load for which “pure” buckling (without eccentricities or lateral
loading) would occur.
The coefficient 0,4 (or 0,8) for estimating the stiffness (see H.1.2 (3)) can be compared to
0,3/(1+¢ef) in expression (5.26). Expression (5.26) is valid for isolated members, where all the
vertical load considered acts on the member itself. Then there effects not only of cracking, but also
of nonlinearity in compression are considered. The last effect can be strong, particularly in cases
where the section is uncracked, usually associated with high vertical load. For the same reason, a
higher stiffness value for uncracked section is not given in 5.8.7.2. In a structure, on the other hand,
most of the vertical load is normally on the braced units, which means that there is less effect of
compression nonlinearity on the bracing units, in which case a particular value for uncracked
section (0,8) is justified
1
. A further difference is that the bending moment normally has a more
favourable distribution in a bracing unit than in isolated members, which gives less overall effect of
cracking. These circumstances together justify the use of 0,4/0,8 instead of 0,3/(1+˻ef). Creep is
not included in the criterion for neglecting second order effects in structures (as it is for isolated
members). The reason is that for global second order effects in structures, the dominating first order
effect is wind. In this circumstance, there is little effect of creep, and consequently, the effective
creep ratio according to 5.8.4 will be low.
The coefficient ç0 in expression (36) depends on various parameters. For constant stiffness, equal
load increment per storey and rigid moment restraint at the base, ç0 will depend on the number of
storeys and (to some extent) on the distribution of vertical load between braced and bracing
members according to Figure 5.16 (the buckling load has been calculated numerically by Vianello’s
method, and ç0 has then been evaluated according to expression (36)).
Figure 5.16. Global buckling due to bending and coefficient for buckling load. Constant stiffness and equal increment of
vertical load per storey.
The coefficient ç0 according to the upper curve in Figure 5.16 can be approximated by
s
0
s
n
7,8
n +1,6
ç ~
where ns = number of storeys
Combining expressions (34) to (36) gives
cd c s cd c cd c
V 0
2 2 2
s
0,4 E I n E I E I
F 0,1 = 0,312 = ȕ
n +1,6 L L L
s ç (38)
where 0,312 = 7,8ā 0,1ā 0,4 and ȕ = 0,312ā ns /(ns + 1,6)
1
The ratio 0,5 between the stiffnesses for cracked and uncracked sections is of course a rough simplification.
The ratio should depend on the reinforcement and the normal force, and with a normal force there is a more or less
smooth transition between the two stages. However, since this is about cases where second order effects are more or
less negligible, a simple rule is acceptable.
Guide to EC2 Section 5
Page 516 Table of contents
This is the background to expression (5.18). Compare the ENV formulation (see 3.3.1 above):
V cm c
L = F E I s o (310)
Expression (5.18) can be formulated in the same way (substituting Ecd with Ecm and explicitly
including partial factor ¸cE = 1,2 on the right hand side):
V cm c
L = F E I ȕ 1,2 s (311)
In Figure 5.17 the two corresponding parameters ȕ/1,2 (EN) and o
2
(ENV) are compared.
For the EN, curves for both cracked and uncracked sections are shown. The ENV gives no values
for cracked section, therefore there is no comparison for this case.
The comparison shows that for uncracked section, the two models give rather similar results,
although the ENV is often much more conservative.
Figure 5.17. Comparison between EN and ENV criteria
5.8.3.3.3 Effect of flexible moment restraint
Flexible moment restraint at the base will reduce the buckling load. This effect can be directly
included in the ç–factor for the buckling load. (For isolated members this effect is normally taken
into account by increasing the effective length. In the global analysis of structures, however, a direct
reduction of the buckling load is more convenient, since the effective length is not a practical
parameter for bracing members and systems having varying axial load).
For bracing units with two or more storeys, reasonably equal increment of the vertical load per
storey and constant cross section, the isolated effect of flexible end restraint can be accurately
modelled by the factor
1
1 0,7
1
k
ç ~
+
(312)
where
M
k =
L EI
u
(same definition as for isolated members, see 3.2)
u = rotation for bending moment M (compare figure 5.13)
The factor ç1 is an approximation, which has been derived by calibration against accurate numerical
calculation for different numbers of storeys, see figure 5.18. The product ç0ç1 corresponds to ç in
expression (H.2) in EN annex H.
Figure 5.18. Effect of flexibility of end restraint for bracing units. Solid curves represent the “exact” solution, dashed
curve the approximation according to expression (312)
The effect of flexible moment restraint is not covered in the ENV, therefore no comparison can be
made.
Effect of global shear deformations
a) Shear deformations only (see Figure 5.19):
The (hypothetical) buckling load for shear deformations only is:
FV,BS = S (or ES for more than one bracing unit) (313)
Here S is the shear stiffness (= shear force giving a shear angle = 1; see figure H1 in the EN).
Guide to EC2 Section 5
Page 517 Table of contents
5.8.4 Creep
Figure 5.19. Hypothetical buckling due to global shear deformations only
b) Combined bending and shear:
The combined buckling load, taking into account bending and shear deformations, can be
expressed as
,
V,BB V,BB
VB
V,BB V,BS V,BB V BS V,BB
F F 1
F = =
1 F +1 F 1+F F 1+F S
~
_
(314)
Expression (314) can be derived analytically for simple cases like isolated members with constant
normal force. By numerical calculations, it can be verified also for bracing units with vertically
increasing axial load and significant global shear deformations (e.g. shear walls with large
openings).
The basic criterion for neglecting second order effects is the same as before:
FV 0,1ā FV,B (315)
which leads to expression (H.6) in Annex H.
This case is not covered in the ENV, therefore no comparison can be made.
C5.8.4. Effective creep ratio
5.8.4.1 General
The ENV stated that creep should be considered in connection with second order effects, but gave
no information on how. In the EN, on the other hand, practical models for taking into account creep
are given, based on the so called “effective creep ratio”.
A general approach would be to first calculate creep deformations under longterm load, then to
analyse the structure for the additional load up to design load. With the effective creep ratio, the
analysis can instead be made directly for the design load in one step.
Figure 5.20 illustrates a hypothetical load history and the corresponding deformations. The total
load is assumed to consist of one Longterm part QL (corresponding to the quasipermanent
combination) and one additional shortterm part up to the Design load QD, applied after a “long
time”.
2
The total load history can then be divided into three parts:
1. AB  longterm load QL giving an elastic deformation
2. BC  constant load QL giving a creep deformation based on full creep coefficient ¢
3. CD  additional load (QD  QL) giving an additional elastic deformation
Figure 5.20. Illustration of load history and deformations
The total deformation under longterm load can also be calculated directly using an equivalent E
modulus
3
for the concrete, Ee = Ec/(1+¢). This corresponds to line AC in figure 5.20.
4
The total deformation under design load can be calculated in a similar way if an effective creep ratio
¢ef is used, line AD in figure 5.20. The “effective equivalent concrete modulus” would then be
Eef = Ec/(1+ ¢ef) where ¢ef is the effective creep ratio.
5.8.4.2 Effect of creep in cross sections
In the following, three examples are used to derive and illustrate the effective creep ratio ĳef.
The examples deal with bending moment and curvature in the following cases, assuming linear
elastic material behaviour:
a) uncracked unreinforced cross section (5.8.4.2.1)
b) uncracked reinforced section (5.8.4.2.2)
c) cracked reinforced section (5.8.4.2.3)
Guide to EC2 Section 5
Page 518 Table of contents
5.8.4.2.1 Uncracked unreinforced cross section
This is the simplest case for demonstrating the idea behind the effective creep ratio. The total
curvature under a longterm bending moment ML is (cf. line AC in figure 5.20):
( )
L
c c L
M 1
= 1+
r E I
 
¢

\ .
(41)
The part caused by creep can be separated:
L
c c C
M 1
=
r E I
 
¢

\ .
(42)
Under design load with total bending moment MD, part of which is a longterm moment ML
with a load history according to figure 5.20, the total curvature will be
( ) 1 1
D D L D L D
ef
c c c c c c c c D c c D C
M M M M M M 1 1
= = = =
r E I r E I E I E I M E I
 
   
+ +¢ +¢ +¢
  
\ . \ .
\ .
(43)
Thus, the effective creep ratio is (cf. expression (5.19), with different notation):
L
ef
D
M
=
M
¢ ¢ (44)
5.8.4.2.2 Uncracked reinforced cross section
The total curvature under longterm bending moment ML can be expressed in the following
simplified way, using an equivalent Emodulus for concrete (see second footnote in 5.20):
1
1
L L
c
L
c s s
M M
E
r EI
I E I
 
= =

\ .
+
+¢
(45)
The part of this curvature caused by creep can be separated:
1
1
L L L
c
c c s s C
c s s
M M M
E
r EI E I E I
I E I
 
= = ÷

+
\ .
+
+¢
(46)
Introducing the following parameters:
ȕ = Įȡ(is/ic)
2
oȡ = (Es/Ec)(As/Ac)
ic = radius of gyration of concrete area
is = radius of gyration of reinforcement area
curvatures can be expressed in the following way:
( ) 1
L
c c L
M 1 1+
=
r E I + 1+ ȕ
¢  

¢
\ .
(47)
( )
L
c c C
M 1 1+ 1
= 
r E I 1+ 1+ ȕ 1+ ȕ
 
¢  



¢
\ .
\ .
(48)
Under design load and total bending moment MD the total curvature will be, including creep due to a
longterm bending moment ML:
( )
D D L
c c c c c c D C
M M M 1 1 1 1 1+ 1
= + = + 
r E I 1+ ȕ r E I 1+ ȕ E I 1+ 1+ ȕ 1+ ȕ
 
¢    

 

¢
\ . \ .
\ .
(49)
The same curvature expressed with the effective creep ratio would be, see expression (47):
( )
ef D
c c ef D
1+ M 1
=
r E I 1+ 1+ ȕ
¢  

¢
\ .
(410)
Combining expressions (49) and (410), after simplification:
( ) ( )
ef L
ef D
1+ M 1 1+ 1
= + 
1+ 1+ ȕ 1+ ȕ M 1+ 1+ ȕ 1+ ȕ
 
¢ ¢


¢ ¢
\ .
(411)
From this the effective creep ratio can be solved:
( )
ef
A 1+ 1
=
1 A
oµ
¢
oµ
(412)
2
Subscripts L and D are used in this chapter for simplicity; they correspond to Eqp and Ed in 5.8.4.
3
Nonlinear effect will be dealt with later, see clause 4.3.
4
Theoretically this is not fully correct, since concrete stresses will decrease and reinforcement stresses increase with
time. However, it is a reasonable approximation in most cases.
Guide to EC2 Section 5
Page 519 Table of contents
where
( )
L
D
M 1 1+ 1
A = + 
1+ ȕ M 1+ 1+ ȕ 1+ ȕ
 
¢


¢
\ .
Figure 5.21 shows the relationship between ¢ef and ML/MD for ¢ = 3 and different values of ȡ
for the uncracked reinforced cross section.
Figure 5.21. Effective creep ratio as a function of the ratio ML/MD for ĳ = 3, uncracked rectangular cross section with
symmetric reinforcement, edge distance t = 0,1h and o =Es/Ec = 6
The straight line for ȡ = 0 is identical with expressions (44) and (5.19). With reinforcement, i.e. ȡ >
0, expression (5.19) becomes more or less conservative.
5.8.4.2.3 Cracked reinforced cross section
The cracked cross section can be treated analogously, although it is a little more complicated.
As the simplest case, consider a rectangular cross section with bending moment only and tensile
reinforcement only. The flexural stiffness in the cracked stage (ignoring any contributions from
concrete in tension) can then be expressed as
( )( )
2
1 1 3
s s
EI E A d = ÷ç ÷ç (413)
where As = area of reinforcement
d = effective depth
ç = x/d
x = depth of compression zone
The relative depth of compression zone for a certain creep coefficient ¢ can be obtained from
( )
( )
2
= 1+ ȡ 1+ 1
1+ ȡ
¢
 
ç ¢ o 

¢ o
\ .
(414)
where oȡ = (Es/Ec)(As/Ac)
To simplify expressions, introduce the symbol
( )( )
1
B =
1 1 3
¢
¢ ¢
ç ç
(415)
The total curvature under design load can then be written:
( )
0 0
2 2 2
D L D
ef
D s s s s s s
M M M 1
= B + B  B = B
r E A d E A d E A d
¢ ¢
 

\ .
(416)
where B0 is parameter according to expression (415) for ¢ = 0 and B˻ef is the same for ¢ = ¢ef.
Values of ¢ef for which expression (416) is satisfied can be found by iteration (direct solution is not
possible in this case). Figure 5.22 shows the result for ¢ = 3.
Guide to EC2 Section 5
Page 520 Table of contents
Figure 5.22. Effective creep ratio as a function of ratio ML/MD for a cracked rectangular cross section with tensile
reinforcement only, based on d = 0,9h and o = 6. Basic creep coefficient ¢ = 3
In this case the curves will approach the straight line according to expression (5.19) the higher the
reinforcement ratio is. However, curves for low and moderate ratios are also quite close.
5.8.4.2.4 Conclusions concerning cross sections
The idea behind the effective creep ratio in 5.8.4 is illustrated in figure 5.20 and demonstrated in
three examples. The simple linear relationship according to expression (5.19) is always more or
less conservative, but deviations are generally small. Furthermore, in a reinforced section the
overall effect of creep on stiffness is reduced with increasing reinforcement, since creep only
affects the concrete contribution to the stiffness. Therefore, the effect of deviations on the stiffness
will not be as high as it may appear from the above figures.
5.8.4.3 Effect of creep in slender columns
5.8.4.3.1 General
The above conclusion concern only cross sections and are based on linear material behaviour.
In this point the relevance of the effective creep ratio for slender columns are examined.
A slender column behaves in a nonlinear way, due to both material and geometrical nonlinearity.
A nonlinear behaviour similar to the linear one in figure 5.20 is outlined in figure 5.23.
Figure 5.23. Illustration of load history and deformations with nonlinear behaviour
The load history can be divided into three steps:
1. Application of longterm load QL, immediate deformation y1, calculated for ¢ef = 0
2. Longterm load QL during time tt0, total deformation y2, calculated for ¢ef = ¢
3. Load increase up to design load QD, additional deformation y3  y2, calculated for ¢ef = 0
A realistic calculation representing this load history should involve these three steps, including the
relevant first order moments or eccentricities for each step. As a simplification, steps 1 and 2 can be
combined into one, using a stressstrain diagram with the strains multiplied by (1+¢), see 6.4. This
corresponds to line AC, and the calculation is then reduced to two steps.
The last step can be calculated in two alternative ways:
a. After calculating point C, the additional load QD – QL is added, with deformation starting from y2.
See line CD in figure 5.23.
b. After calculating point C, the total load QD is applied “from scratch”, but with y0 = y2  y1 as an
initial deflection added to other first order effects. See line ED in figure 5.23.
Alternative b. will be used in twostep calculations in the following way. The distribution of y0 along
the column should in principle be the same as the distribution of y2 – y1. For a pinended column,
Guide to EC2 Section 5
Page 521 Table of contents
however, a sineshaped or parabolic distribution will be adopted as a simplification.
A further simplification is a onestep calculation, using an effective creep ratio ¢ef; (line AD in the
figure). For the definition of ĳef there are two main options:
a) based on first order moments M0L and M0D, i.e. ¢ef = ¢ā M0L / M0D
b) based on total moments ML and MD, including 2nd order moments, i.e. ¢ef = ¢ā ML / MD
The relevant deformation parameter in second order analysis is curvature, which depends primarily
on bending moment. Therefore, the axial load should not be included in the definition of effective
creep ratio.
Alternative b) is the most realistic one, since creep deformations will mainly be governed by total
moments. With this alternative, however, iteration is inevitable since second order moments depend
on stiffness, which depends on effective creep ratio, which depends on total moments etc.
Therefore, alternative a) will be the normal choice in practical design.
Alternative a. is always more or less on the safe side. The reason is that the second order moment
is a nonlinear function of the axial load. Therefore, the moment increase due to second order
effects will be greater under design load than under longterm load, and the ratio ML/MD will be lower
if second order moments are included. This is easy to verify with a magnification factor based on
linear material behaviour (see chapter 5.8.7); this tendency will be even stronger in a nonlinear
analysis.
5
5.8.4.3.2 Comparison between one and twostep calculations
An example will be used to compare the onestep calculation, using the effective creep ratio, with
the more realistic twostep calculation. A high slenderness ratio has been chosen, in order to
emphasize the effects considered. All calculations below have been done with the general method.
(for a general description and discussion of this method, see chapter 5.8.6). Geometric assumption
are:
 Concrete C40
 Reinforcement S500
 Rectangular cross section with reinforcement concentrated to opposite sides
 Mechanical reinforcement ratio e = 0,15 (total reinforcement)
 Edge distance of reinforcement 0,1h
 Eccentricity e0 = 0,08h (same for longterm and design load; no other first order effect)
 Slenderness l/h = 40 (ì= 139)
 Basic creep coefficient ¢ = 3
In the following, all axial loads and bending moments are expressed in relative terms, i.e.
n = N / Acfcd and m = M / hAcfcd. Therefore, no absolute dimensions are used.
a): nL = 0,100 (longterm axial load)
1. immediate deformation, calculated with ¢= 0: y1/h = 0,0173
2. total deformation, calculated with ¢ = 3: y2/h = 0,0819
3. creep deformation: y0/h = 0,0819 – 0,0173 = 0,0646
y0 is taken as an initial deflection with parabolic distribution, and is added to the constant first order
eccentricity e0 given above. The load capacity under this total first order effect and no creep (¢ = 0)
is calculated. The result is nRd = 0,235
b): nL = 0,125
1. y1/h = 0,023 (¢ = 0)
2. y2/h = 0,134 (¢ = 3)
3. y0/h = 0,134 – 0,023 = 0,111 (creep deformation)
The load capacity calculated with e0+ y0 and with ¢ = 0 is nRd = 0,189
These values are compared to the result of a onestep calculation, using an effective creep ratio
based on first order moments.
a) nL = 0,100
¢ef = ¢ā M0L / M0D = ¢ā NLā e0 / (NDā e0) = ¢ā nL / nD = 3 0,100/0,235 = 1,28
6
The load capacity with ¢ef = 1,28 is nRd = 0,198
Cf. 0,235 in twostep calculation; thus the result is 16% conservative
5
A "curvature method", giving a fixed 2nd order moment, would lead to the wrong conclusion here.
6
In this particular example the first order moment is proportional to the axial load, therefore the effective creep ratio can
be based on axial loads as well as moments. In the general case only moments should be used.
Guide to EC2 Section 5
Page 522 Table of contents
b) nL = 0,125
¢ef = 3ā 0,125/0,189 = 1,98 ĺ nRd = 0,166
Cf. 0,189 in twostep calculation; thus the result is 12 % conservative
These results are somewhat conservative, as could be expected (the reason is explained above).
Next is a onestep calculation with ¢ef based on total moments.
a) nL = 0,100
Total moment under nL is mL = nLā (e0 + y2) = 0,100ā (0,08 + 0,0819) = 0,0162
After iteration the following values are found:
Total moment under design load mD = 0,0618
Effective creep ratio ¢ef = ¢ā mL / mD = 3ā 0,0162/0,0618 = 0,786
Load capacity with ¢ef = 0,786 is nRd = 0,224 (total moment for this load is mD = 0,0618)
This is within 5 % of the twostep calculation (which gave nRd = 0,235)
b) nL = 0,125
Total moment under nL is mL = nLā (e0 + y2) = 0,125ā (0,08 + 0,134) = 0,0267
After iteration: mD = 0,0531, ¢ef = ¢ā mL / mD = 3ā 0,0267/0,0531 = 1,151, nRd = 0,183
This is within 3 % of the twostep calculation (nRd = 0,189)
5.8.4.3.3 Conclusions
It is conservative to use an effective creep ratio based on first order moments; total moments will
give more accurate results. In practical design, however, total moments are much more complicated
to use, however, since iteration will be necessary. Therefore, the normal procedure will be to use
first order moments. This is further discussed below.
5.8.4.4 The effect of creep on slenderness limit
The effect of creep on slenderness limit will be further studied here, comparing the onestep and
twosteps methods according to 4.3.1 and 4.3.2. It is thus a complement to clause 3.1, dealing with
the slenderness limit in general. It is also a complement to 4.3.2, dealing with creep combined with
a high slenderness, since this clause deals with low slenderness ratios.
Table 5.7 shows the results of calculations, based on a slenderness corresponding to the limit for
which second order effects may be neglected with ¢ef = 0, see 3.1. The basic parameters are the
same as for the example in 4.3.2, except those for which different values are given.
Table 5.7. The effect of creep for columns with a low slenderness.
Symbols in table:
e02 the greater of the two first order eccentricities
e01 the lesser eccentricity
l0/h slenderness corresponding to the limit for 10 % moment increase at ¢ = 0
n relative normal force N/Acfcd
nu0 load capacity for the current slenderness and ¢ = 0
nL longterm load
¢ef effective creep ratio = ¢ā nL / n0; here ¢ = 3 has been assumed
Guide to EC2 Section 5
Page 523 Table of contents
nu1 load capacity including the effect of creep according to 1 step method
nu2 load capacity including the effect of creep according to 2 step method
Comments
The agreement between the 1step and 2steps methods is in most cases good. For e01/e02 = 0,9
(double curvature bending) the 1step method is generally slightly conservative compared to the 2
steps method. There are also a few cases where the opposite is true, but in these cases the long
term load is close to the limit where instability would occur with ¢ef = ¢, and then the 2steps
method becomes uncertain. In these cases, the result would have been more representative with a
somewhat lower longterm load.
The use of an extended stressstrain diagram in the 2steps method can be discussed. In principle it
means that creep deformations will correspond to the stresses in the final stage. In a more accurate
calculation they should be integrated from 0 to ¢, with increasing second order moment. However,
the error will be small, since the stresses are normally not very high under longterm load, and since
second order moments are small at these low slenderness ratios.
In most cases a first order eccentricity e02/h = 0,32 has been used, with the aim of having a
moderate normal force. For the sake of completeness, one case with a high normal force is also
included (nu0 = 1,022, e02/h = 0,08) and one with a low normal force (nu0 = 0,122, e02/h = 1,28).
Even with the low normal force, there is a significant effect of creep. (The 2steps method in this
case seems to give more effect of creep for low than for high normal force. This is misleading,
however; it is a consequence of the longterm load being close to the instability load for ¢ef = ¢, see
discussion in the first paragraph above.)
Conclusion concerning the effect of creep on the slenderness limit
In these examples, creep reduces the load capacity by 5 to 30% (average 15%). If second order
moments are neglected, which is allowable at these slenderness ratios, the result is in principle
already 10% on the unsafe side. If creep would also be neglected, the results would be another 5 to
30% on the unsafe side.
The conclusion is that creep can not be neglected in the slenderness limit.
5.8.4.5 Safety under longterm load only
The effective creep ratio is based on moments under quasipermanent load which, according to its
definition in EN 1990, is an SLS load with no load factors (except ¢2 < 1 for variable loads). Thus, in
the extreme case of permanent load only, assuming first order moments proportional to the load as
in the above examples, the highest possible effective creep ratio is
¢ef = ¢ā M0L/M0D = ¢ā 1,0/1,35 = 0,74¢
The following question now arises:
Can load NL together with ¢ef = ¢ be more severe than ND = 1,35ā NL with ¢ef = 0,74¢ ?
The example in 4.3.2 is used again. The load capacity for ĳef = 0,74¢ = 2,22 is found to be nRd =
0,159. The corresponding longterm load is nL = 0,159 / 1,35 = 0,118. The load capacity for full
creep, ¢ef = ¢ = 3, is found to be nRdL = 0,141. This is higher than the current long term load, and
the “safety factor” is
¸L = 0,141/0,118 = 1,19
This safety factor may be considered somewhat low, although it should be observed that it is not the
whole safety factor as the normal material safety factors are already included in the calculated
capacities. A reasonable lower limit for the load safety factor could be 1,35.
As shown above, it is conservative to use an effective creep ratio based on first order moments.
The “extra” safety can be estimated by comparison with more accurate calculations, e.g. a twostep
calculation or a onestep calculation with ¢ef based on total moments.
A twostep calculation according to the above scheme is done with different values of nL, until a
value is found for which nRd = 0,159. This happens for nL = 0,134. Thus, one could say that the
additional “builtin” safety is 0,134/0,118 = 1,14, and the total safety against creep failure would be
¸L = 1,14ā 1,19 = 1,36
This is considered to be sufficient, and it can be shown that this factor will be higher for lower values
of slenderness, higher first order moments and higher amounts of reinforcement. In this respect, the
current example is rather extreme, in the unfavourable direction.
Furthermore, longterm load = 74 % of design load is the worst possible case for consideration of
the effect of creep. In normal cases there is always some variable load. The percentage of longterm
Guide to EC2 Section 5
Page 524 Table of contents
5.8.5 Methods of analysis
5.8.6 General method
load then decreases, since variable loads are included in QL with ¢2ā Qk, where ¢2 < 1
8
, whereas in
QD they are included with ¸Qā Qk, where ¸Q > 1. Therefore, the more variable the load, the higher the
safety against “creep failure” will be.
The conclusion is that a onestep calculation, using an effective creep ratio based on first order
moments, will give sufficient safety against failure under quasipermanent load with full creep.
Therefore, this case need not be checked separately, and it is not necessary to include any safety
factor on M0L in the definition of ¢ef.
In 5.8.4 (3) the alternative of using total moments in the definition of effective creep ratio is given.
This is less conservative, however, and most of the extra safety against “creep failure” is then lost.
Therefore, 5.8.4 (3) states that a separate check should then be made for 1,35 QL and with ¢ef = ¢.
This may become the governing factor in cases where the percentage of longterm moment is
moderate or high, more precisely when first order moment ratio M0L / M0D > 0,5.
C5.8.5. Methods of analysis
Three basic methods are described in 5.8.5. Of the simplified methods, (b) is basically the “model
column method” in the ENV, with some modifications. The old name is not used here, since it tells
nothing about the method (all methods are based on models). A more suitable name is “curvature
method”, since the method is based on the estimation of a curvature.
This name will be used here, together with “stiffness method” for method (a), which is based on the
estimation of stiffness.
There are simplified methods other than those mentioned in EC2. One such method, combining
analysis and cross section design in one step, will be shortly described here as an example (it is
currently used in the Swedish code).
It can be used for isolated columns with formally centric load, i.e. no other first order effect than the
prescribed imperfection. The load bearing capacity is given as
NRd = kcfcdAc + ksfydAs (51)
where kc and ks are coefficients depending on slenderness ratio, imperfection, concrete grade,
effective creep ratio etc, calibrated against calculations with the general method.
A method of this type works ideally if the imperfection, an eccentricity or an initial deflection, is
proportional to the buckling length of the column. This is the case in some codes, but not in the
Eurocodes. If the imperfection is proportional to the effective length, the coefficients can be given in
one simple table or diagram with slenderness as the basic parameter.
If the imperfection is not proportional to the effective length, then the absolute value of this length
must be added as a separate parameter, which complicates the presentation (for example, one
diagram or table would only be valid for one length). However, with some simplifications this type of
method could be useful also under EN 1992, particularly for storey high pinended columns, which
are common as interior columns in buildings.
If there are first order moments other than that due to the imperfection, a separate design for normal
force and (magnified) moment must be made. A special moment magnification factor is included in
the method for such cases, but the simplicity is lost and the method no longer has any particular
advantages over the “stiffness” or “curvature” methods in EC2.
In the following chapters, the general method and the simplified methods (a) and (b) are described.
C5.8.6. General method
5.8.6.1 General
The most accurate of the methods described in 5.8.5 is the ”general method”. It is based on non
linear analysis, including both material and geometric nonlinearity (second order effects).
”General” here refers to the fact that the method can be used for any type of cross section, any
variation of cross section, axial load and first order moment, any boundary conditions, any stress
strain relations, uniaxial or biaxial bending etc. The limiting factor is the capability of the available
computer program. The method rests on a few simple assumptions:
• linear strain distribution
• equal strains in reinforcement and concrete at the same level
• stressstrain relationships for concrete and steel
8
EN 1990 gives values for ¢2. For some loads, e.g. wind, ¢2 = 0. A common value is 0,3 (office and residential
areas). The highest value given is 0,8.can be based on axial loads as well as moments. In the general case only
moments should be used.
Guide to EC2 Section 5
Page 525 Table of contents
Conditions of equilibrium and deformation compatibility are satisfied in a number of cross sections,
and the deflection is calculated by double integration of the curvature, having an assumed variation
between the selected sections. This may be selfevident, but it is mentioned in 5.8.6 (6) as a
reference for a simplified version, in which only one cross section (or certain critical sections) is
studied, and the curvature is preassumed to have a certain variation in other parts of the member.
This gives simpler computer programs and faster calculation, but less accuracy. See figure 5.24.
Figure 5.24. Illustration of accurate (left) and simplified (right) versions of the general method
Any stressstrain relations can be used. A continuous curve with a descending branch is considered
to be the most realistic alternative for the concrete; it is also convenient for computational reasons.
Creep can be considered in different ways; the simplest way is to multiply all concrete strains by
(1+¢ef), see clause 6.4.
Tension stiffening (i.e. the contribution from concrete in tension between cracks) can easily be
taken into account in the general method, e.g. by using a descending branch of the concrete stress
strain curve in tension, by modifying the stressstrain curve of the reinforcement or by any other
suitable model. In the calculations presented in this report, however, all contributions from concrete
in tension have been ignored; this is always more or less conservative.
5.8.6.2 Safety format
The safety format in nonlinear analysis has been much debated, and different models have been
proposed. The safety format is particularly important in second order analysis, where the absolute
magnitude of deformations has a direct influence on the ultimate load.
9
The safety format should satisfy two basic criteria.
1. It should be possible to use the same set of material parameters in all parts of the member, in
order to avoid discontinuities and computational problems.
The model in ENV 199211 (Appendix 2) does not comply with this, since it assumes mean values
of material parameters for the calculation of deformations and design values for the check of
resistance in critical sections. This also means that there will be no “material safety” at all in the
calculated resistance, in cases where failure occurs before reaching the design cross section
resistance (stability failure) – unless “critical section” is substituted by some “critical length” (which
then remains to be defined, however).
2. The safety format should be compatible with the general design format based on partial safety
factors.
The model in ENV 19922 (Appendix B) does not comply with this, since it uses mean values for the analysis and a
global safety factor ¸R = 1,3 to reduce the ultimate load resulting from the analysis. This gives the same results as using
design values fcm/1,3, fyk/1,3, Ecm/1,3 and Esm/1,3. Thus, it makes no difference whether the ultimate load is governed by
concrete or steel, resistance or stiffness. The reduction of the reinforcement strength is too severe, as is also the
reduction of the material stiffness parameters, particularly for reinforcement (Esm/1,3). A nonlinear analysis using this
safety format will be conservative, and the potential benefits of using a refined method are lost.
The safety format defined in 5.8.6, based on using design values in the analysis, satisfies both
criteria. A design value of the ultimate load will be obtained as a direct result of the analysis, and the
problems associated with the abovementioned safety formats are avoided. Since the Emoduli vary
less than the corresponding strengths, the partial safety factors given for E should be lower than for
f:
9
The absolute magnitude can be of importance also in e.g. continuous beams, but only in the check of rotation
capacity, and it would normally not have the same direct influence on the ultimate load as in 2nd order analysis.
10
This diagram is taken from [1], which primarily deals with high strength concrete according to Swedish rules, but this
makes no difference for what the diagram is intended to show.
Guide to EC2 Section 5
Page 526 Table of contents
For concrete, ¸c = 1,5 for strength takes into account not only strength variation, but also
geometrical deviations in the cross section. Assuming a factor 1,1 for these deviations, and
considering the relationship between strength and Emodulus, a reasonable value of the factor for
Ec is ¸cE = 1,1ā (1,5/1,1)1/3 § 1,2.
For steel, ¸s = 1,15 includes a factor of about 1,05 for geometrical deviations. Thus, a design value
Esd=Esm/1,05 would be logical, considering the fact that variations in the Emodulus are negligible.
However, a factor 1,0 has been chosen as a simplification, and in order not to deviate from 3.2.3;
differences in terms of calculated result are negligible.
5.8.6.3 Interaction diagrams
The resistance of slender columns resulting from a general analysis can be shown in a practical
form with interaction curves, figure 5.25. One such curve shows the maximum first order moment
M0 (or eccentricity e0 = M0/N) for a certain axial load N.
The thin curves in figure 5.25 show the total moment M as a function of N for a given e0. The higher
the slenderness, the more the total moment M increases over the first order moment M0. (Note that
the diagram gives axial load and moment in relative terms n and m.) One point on the interaction
curve for a given slenderness is obtained by plotting the maximum value of n on the line
representing m0 or e0. This is demonstrated in figure 5.25 for one relative eccentricity e0/h = 0,1 and
different slenderness values ì = 35, 70, 105 and 140.
The difference Mu  M0 between the cross section resistance (curve ì = 0) and the first order
moment at maximum load represents the second order moment. However, in some cases there is a
stability failure before any cross section reaches its ultimate moment, and then the “true” second
order moment is less than Mu  M0. This occurs for ì = 105 and 140 in figure 5.25.
This nominal second order moment Mu  M0 is useful as a basis for simplified methods; see clause
6.5 and chapters 7 and 8.
Figure 5.25. Interaction curves for columns of different slenderness, calculated with the general method. Rectangular
cross section. n and m0 are relative axial force and first order moment respectively, i.e. n = N / bhfcd, m0 = M0/bh
2
fcd. All
curves are based on e = 0,2 and ¢ef = 0.
Concrete grade is C80.
10
First order moment is constant, e.g. caused by equal end eccentricities
5.8.6.4 The effect of creep
Creep can be taken into account in different ways. The most accurate model would be to increase
load and time in steps, for each step taking the stresses, strains (and corresponding deflections)
from the previous step as starting values for the next increment. For each step, strains would be
calculated taking into account their timedependence.
A simplified model is to multiply all strain values in the concrete stressstrain function with the factor
(1+¢ef), see figure 5.26, where ¢ef is an effective creep ratio relevant for the load considered. With
this model, the analysis can be made either in steps for loads of different duration, or directly for the
design load combination in one step, see chapter 4.
For creep in slender members in particular, see clause 4.3.
Figure 5.26. Simple way of taking into account creep in general method
Guide to EC2 Section 5
Page 527 Table of contents
Figure 5.27 is calculated in this way, using ¢ef = 2,0 and other parameters the same as in figure
5.25. Curves according to figure 5.25 are also included (dashed), showing the reduction of the load
capacity resulting from creep. The relative reduction increases with slenderness.
Figure 5.27. Interaction curves for ¢ef = 2. Other parameters are the same as in figure 5.25
Dashed curves are the corresponding curves from figure 62, i.e. for ¢ef = 0.
The difference represents the effect of creep
Another question is whether one and the same effective creep ratio should be used along a
compression member (or in different parts of a structure), or if it should vary as the ratio MEqp/MEd
may vary. The latter would be the most correct alternative, but normally it is reasonable to use one
representative value of ¢ef for a member or even a whole structure.
5.8.6.5 Simplified methods and their common basis
In a simplified calculation method one can use the difference between cross section resistance and
first order moment, Mu  M0 in figure 5.25, as a nominal second order moment. When this moment is
added to the first order moment, a design moment is obtained for which the cross section can be
designed with regard to its ultimate resistance. As pointed out above, this nominal second order
moment is sometimes greater than the ”true” second order moment.
However, it can give correct end results, even in cases where the load capacity is governed by a
stability failure before reaching the cross section resistance, if given appropriate values.
For practical design, there are two principal methods to calculate this nominal second order
moment:
1. estimation of the flexural stiffness EI to be used in a linear second order analysis (i.e.
considering geometrical nonlinearity but assuming linear material behaviour); this method is
here called stiffness method, see chapter 5.8.7
2. estimation of the curvature 1/r corresponding to a second order deflection for which the second
order moment is calculated; this method is here called curvature method, see chapter 5.8.8.
Before entering into details of the two methods in chapters 5.8.7 and 5.8.8, their common basis will
be shortly described.
The total moment including second order moment for a simple isolated member is:
2
0 2 0 0
1 l
M M M M N y M N
r c
= + = + = + (61)
where (see figure 5.28)
M = total moment
M0 = first order moment
M2 = second order moment
N = axial force
y = deflection corresponding to 1/r
1/r = curvature corresponding to y
l = length
c = factor for curvature distribution
Guide to EC2 Section 5
Page 528 Table of contents
5.8.7 Method based on
nominal stiffness
5.8.7.1 General
5.8.7.2 Nominal stiffness
5.8.7.3 Moment magnification
factor
Figure 5.28. Illustration of deformations and moments in a pinended column. (In the figure, first order moment is
exemplified as the effect of a transverse load. First order moment could also be given by eccentricity of the axial load.)
The difference between the two methods lies in the formulation of the curvature 1/r.
In the stiffness method 1/r is expressed in terms of an estimated nominal flexural stiffness EI:
1 M
=
r EI
(62)
The stiffness EI should be defined in such a way that ULS cross section design for the total moment
M will give an acceptable end result in comparison with the general method. This includes, among
other things, taking account of cracking, creep and nonlinear material properties.
In the curvature method, the curvature 1/r is estimated directly, on the basis of assuming yield strain
in tensile and compressive reinforcement:
2
1
0,9
yd
=
r d
c
(63)
This model overestimates the curvature in those cases where yielding is not reached, giving a too
conservative end result. The typical example is where the ultimate load is governed by stability
failure, before reaching the cross section resistance. The model may also underestimate the
curvature in some cases, since it does not take into account creep. However, various corrections
can be introduced to improve the result.
In the following chapters the two simplified methods will be described and compared to the general
method.
C5.8.7. Method based on stiffness
5.8.7.1 Basic equations
A simple isolated column is considered, e.g. pinended with a length l = l0; see figure 5.28. The
second order moment can be expressed in the following way, cf. equation (61) and fig. 5.28:
2 2
0 0 2
2
0 2
1 l M M l M
M = N y = N N
r c EI c c c
 
= +

\ .
(71)
With c0 and c2 it is possible to consider different distributions of first and second order moments
(primarily the corresponding curvatures). Solving for M2 gives
2
0
0 2 0
2 0 0
2 2
0 2 0
1
1
0
l
N
c EI c c
M M M
l c EI l N
N
c EI
= =
÷
÷
(72)
In many cases it is reasonable to assume that the second order moment has a sine shaped
distribution. This corresponds to c2 = t
2
, and M2 can then be written
( )
2
2 0 0
2 2
0
/ 0
/ 1
/ 1
B
c
M = M = M
N N 
EI l 
t 
t
(73)
where NB = nominal buckling load (based on nominal stiffness).
ȕ = t
2
/c0, parameter taking into account the distribution of first order moment
The total moment will be
0
1
/ 1
B
M = M +
N N 
  

\ .
(74)
which corresponds to equation (5.28) in 5.8.7.
Guide to EC2 Section 5
Page 529 Table of contents
5.8.7.2 Moment distribution
In some cases the value of c0 is known, as in the examples mentioned in 5.8.7.3 (2).
The case of differing end moments will be examined more closely. A reference is made to 5.8.8.2
(2), with the wellknown formula for an equivalent constant first order moment:
M0e = 0,6 M02 + 0,4 M01 0,4 M02 (75)
This is illustrated in figure 5.29.
Figure 5.29. Illustration of equivalent moments in case of differing end moments
Equation (74) can be used with the equivalent first order moment according to (75) also. An
example of the result is shown in figure 72, where two different c0 values were used: 8 and 10
respectively.
Figure 5.30. Slender member with differing end moments according to figure 71 with e02/h = M02/Nh = 0,1 and NB/N =
2: Comparison between maximum moment according to exact solution and equivalent first order moment (75) with
magnification factor (74).
Thick line = exact solution.
Upper thin line = equivalent moment with c0 = 8
Lower thin line = equivalent moment with c0 = 10
Figure 5.30 shows good agreement with the exact solution for c0 = 8, whereas for c0 = 10 slightly
unsafe results may arise. Therefore c0 = 8 is recommended in 5.8.7.3 (2); this is also consistent with
the assumption of a constant equivalent first order moment. The example is based on a
comparatively high second order effect (N/NB = 0,5), which enhances the differences.
In many cases it is reasonable to assume that first and second order moments have similar
distributions, in which case ȕ § 1. Equation (74) can then be simplified to
0
/ 1
B
M
M =
N N 
(76)
This corresponds to equation (5.30) in 5.8.7.3. It can be shown that this expression can be used
also for structures, provided a global buckling load can be defined. See 5.8.7.4 for global analysis of
structures.
Guide to EC2 Section 5
Page 530 Table of contents
5.8.7.3 Estimation of stiffness
The maximum first order moment M0 for different axial forces, slenderness ratios, and reinforcement
ratios can be determined using the general method, see chapter 5.8.6.
The values thus obtained can be considered the “correct” ones. With the “stiffness” method, on the
other hand, the maximum first order moment can be expressed on the basis of equation (74),
assuming that the total moment is equal to the ultimate moment resistance Mu for the normal force
N:
0
2 2
1 1
1 1
u u
B
M M
M
N N EI l N
= =
 
+ +
÷ t ÷
(77)
The following is a simple model for the stiffness, expressed as the sum of separate contributions
from concrete and reinforcement:
EI=Kc Ec Ic + Ks Es Is (78)
where Ec, Es = concrete and steel Emoduli respectively
Ic, Is = moment of inertia of concrete and steel area
The correction factors Kc and Ks can be calibrated using more or less sophisticated models, to give
the required agreement between expression (77) and the general method. In 5.8.7.2 (2) basically
two alternative models are given: a) (expr (5.22)) is a more accurate alternative, valid for
reinforcement ratios down to ȡ = 0,002. b) (expr (5.26)) is a simplified alternative, valid only for
reinforcement ratios ȡ 0,01. Thus, for ȡ < 0,01 only a) may be used, for ȡ 0,01 either method
may be used.
a) if p 0,002
( )
1 2
1
/ 1
s
c ef
K
K k k
=
= +¢
(79)
b) if p 0,01
( )
1
0,3/ 1 0,5
s
c ef
K
K
=
= + ¢
(710)
where ȡ is the geometrical reinforcement ratio, As/Ac
¢ef is the effective creep ratio, see chapter 4
k1 depends on concrete strength class, see (711)
k2 depends on axial force and slenderness, see (712)
1
/ 20
ck
k = f (711)
2
0,20
170
k = n
ì
s (712)
where n is the relative axial force, NEd / (Acfcd)
Ac is the area of concrete cross section
ì is the slenderness ratio, l0/i
For cases where ì is not defined, a simplified alternative to (712) is also given (5.25):
k2 = nā 0,30 0,20 (713)
More sophisticated models for estimating the stiffness can be found in [2] and [3]. Background, see [1].
The results of calculations with stiffness evaluated according to expressions (79) to (712) are
presented in Appendix 2 of this report, in the form of comparison with calculations done using the
general method. The Appendix also compares the curvature method; see chapter 8.
5.8.7.4 Linear analysis of structures
Clause 5.8.7 opens the possibility of using linear second order analysis for structures, using
reduced stiffness(es) taking into account the effect of cracking, creep and material nonlinearity in a
simplified way. Without this possibility, the only alternative for second order analysis of structures
would be nonlinear analysis.
When global second order effects are significant, the effects of cracking etc. may be as important as
for isolated members. It should also be kept in mind that second order effects may be significant in
a structure, even if the geometrical slenderness of individual bracing units is small, in case the
braced units carry a comparatively high vertical load.
The paragraphs applicable to structures are 5.8.3.3 (criterion for ignoring global second order
effects), 5.8.7.3 (3) and Annex H. Two different approaches can be distinguished, one based on a
magnification factor for bending moments, 5.8.7.3 (3), and the other one based on a similar factor
for horizontal forces, H.2.
Guide to EC2 Section 5
Page 531 Table of contents
The two approaches are basically the same, but the one based on moments is suitable mainly for
structures with bracing units consisting of shear walls without significant global shear deformations,
or structures braced by simple cantilever columns, see examples in figure 5.31.
Figure 5.31. Example of structures where a magnification factor can be applied directly to bending moment(s)
in bracing unit(s)
The approach based on magnification of horizontal forces, on the other hand, can be used for all
kinds of structures, and it should be used for frames, shear walls with large openings etc. If properly
used, it gives the correct second order effects in structural systems like frames, shear walls with or
without openings etc; see the schematic example in figure 5.32.
Figure 5.32. Example of a structure where the magnification factor should be applied to horizontal forces rather than to
bending moments. (No deformation is shown in this case, second order effects are instead assumed to be included in
the fictitious, magnified horizontal force HEd.)
Expressions (5.30) and (H.7) are useful if the global buckling load can be defined without difficulty,
like in certain regular structures, see e.g. 5.8.3.3 and H.1. In other cases second order effects may
be calculated stepwise as indicated for a simple frame in figure 5.33.
Figure 5.33. Illustration of stepwise calculation of second order effects.
a) Horizontal load H0 (without vertical load) gives deformation y0.
b) Vertical load V on deformed structure gives additional deformation y1.
c) H1 is an equivalent horizontal load that would give the same deformation y1.
d) Vertical load V and deformation y1 give additional deformation y2.
e) H2 is equivalent horizontal load giving the same deformation y2 etc ....
The total equivalent horizontal force is
H = H0 + H1 + H2 + H3 +… (714)
If ki = Hi/Hi1 is < 1, then the sum H will be finite (i.e. the structure is stable). With increasing number
Guide to EC2 Section 5
Page 532 Table of contents
5.8.8 Method based on
nominal curvature
of steps, in a linear analysis, the ratio ki will sooner or later become constant. In other words, the
following terms will form a geometric series.
The simplest alternative is to assume that all terms, including H0, will form a geometric series.
The total equivalent horizontal force is then obtained as
0 0
1 1 0
1 1 /
H H
H = =
 k  H H
(715)
which is equivalent to expression (H.8).
Note. Expression (715), including the definition of k, can also be expressed in terms of y or a
relevant M.
It can also be shown that the final value of k is equal to the ratio V/VB, where V is the total vertical
load and VB is the global buckling load. Thus, the method of stepwise calculation can be seen as
just a different formulation or derivation of the method based on a magnification factor.
If the distribution of H1 is significantly different from that of H0, the accuracy can be improved by
including one or more steps:
1 2
2 1 3 2
, ... .
1 / 1 /
0 0 1
H H
H = H + H = H +H + etc
 H H  H H
(716)
The simple alternative (715)/(H.8) is sufficiently accurate in most cases, compared to other
uncertainties like the effect of stiffness variations within and between members due to cracking etc.
It should be observed that variation in the degree of cracking between first and following steps does
not have to be considered, if reduced stiffness values according to 5.8.7.2 are used; these values
are intended to be valid for the final stage of deformation. However, if values for uncracked section
are used in early steps, although cracking might occur in later steps, then more steps have to be
included in the analysis, like in expression (716); kvalues for early steps would otherwise be too
low for later steps. This would apply generally when a more refined analysis is used, where gradual
cracking is taken into account.
When the structure is analysed for the equivalent horizontal force HEd, the relevant second order
effects can be obtained everywhere in the structure.
To magnify all moments with the same factor, as in expression (76)/(5.30), would not be correct in
for instance a frame or a shear wall with large openings.
C5.8.8. Method based on curvature
Basic relationships
This method is basically the same as the previously socalled “model column” method in the ENV.
The second order moment is expressed in the following way, cf. equation (61):
2
0
2
1 l
M = N y = N
r c
(81)
As mentioned in 6.5, 1/r is estimated on the basis of reaching yield strain in tensile and compressive
reinforcement. Here correction factors Kr and K
¢
are included:
0
2
1 1
0,9
yd
r r
= K K = K K
r r d
¢ ¢
c
(82)
In chapter 5.8.5 an extended definition of the effective depth d has been introduced, in order to
cover cases where there is no unambiguous definition of d; see figure 5.34 where is is the radius of
gyration of the total reinforcement area
Figure 5.34.Effective depth in cross sections with reinforcement distributed in direction of bending
In order to reduce the curvature in cases where yielding is not reached in the tensile reinforcement,
a factor Kr is introduced (same as K2 in ENV 199211, 4.3.5.6.3):
Kr = (nu  n) / (nu  nbal) 1 (83)
Guide to EC2 Section 5
Page 533 Table of contents
5.8.9 Biaxial bending
where n = NEd / (Ac fcd), relative normal force (Ȟ in the ENV)
NEd is design value of normal force
nu = 1 + e
nbal is value of n at maximum moment resistance; the value 0,4 may be used
e = As fyd / (Ac fcd)
As is total area of reinforcement
Ac is area of concrete cross section
There is another factor K1 in ENV 199211, 4.3.5.6.3 (2), which reduces the curvature for values of
ì between 15 and 35. The purpose of this factor was presumably to avoid discontinuity in cases
where second order effects may be ignored. However, second order effects will often be ignored for
ì between 25 and 35 (see 5.8.3.2), so discontinuities will still occur.
Furthermore, independent of method, there is always a basic discontinuity following from the rule
that second order effects may be ignored if they are below a certain limit. For these reasons, the
factor K1 has not been included in 5.8.8.
The ENV gives no indication of how to take into account creep in the “model column” method.
Comparisons with the general method indicate that in certain cases the method can give unsafe
results if allowance for creep is not considered, and the factor K
¢
has been introduced for this
purpose. It has been calibrated against calculations with the general method.
More sophisticated models for estimating the curvature can be found in [2] and [3]. Their
background is presented in [1].
Comparison with general method and stiffness method
The result of calculations with curvature according to expressions (81) to (83) is presented in
Appendix 2, in comparison with calculations based on the general method. In the same Appendix
calculations with the stiffness method (chapter 7) are also presented and compared.
Using the curvature method for structures
In 5.8.5 there is an indication that the curvature method can be used also for structures, “with
proper assumptions concerning the distribution of curvature”. This statement is based on [4], where
a method is given by which the curvature method can be used also for second order analysis and
design of unbraced frames. For details, see [4].
C5.8.9. Biaxial bending
The general method is suitable for biaxial bending also. The same principles as in uniaxial bending
apply, although the complexity of the problem increases.
Simplified methods like the stiffness or curvature method can also be used. They are then used
separately for each direction, and if the resulting bending moments fulfil a certain criterion, given in
expression (5.38), no further action is necessary.
The criterion in (5.38) is similar to expressions (4.74) and (4.75) in the ENV, 4.3.5.6.4, but there is
one important difference: the ENV check concerns only first order eccentricities, whereas in 5.8.9 it
concerns total eccentricities including second order effects. The reason for including the second
order effects is illustrated in figure 5.35:
Figure 5.35. Example of member with different slenderness in the two directions
Assume for example ì = 100 in one direction and ì = 20 in the other. Second order effects will then
be significant in one direction but negligible in the other. A and B are two examples of the position
of the axial load, both fulfilling the criterion for separate checks according to the ENV, based on first
order eccentricities. This would be acceptable for case A, since the second order effect will make
the total eccentricities even “less biaxial”. It is not acceptable for case B, however, since the second
order effect will now give total eccentricities outside the “permissible” area. Thus, a first order
criterion can be misleading and unsafe.
If criterion (5.38) is not fulfilled, the cross section should be designed for biaxial bending. A simple
Guide to EC2 Section 5
Page 534 Table of contents
5.9 Lateral instability of
slender beams
5.10 Prestressed members
and structures
5.11 Analysis for some
particular structural members
model for this, “in the absence of an accurate cross section analysis”, is given in 6.1:
a
a
y
x
Rx Ry
M
M
1
M M
 
 
+ s 


\ .
\ .
(91)
where Mx/y design moment in the respective direction, including nominal 2nd order moment
MRx/y corresponding moment resistance of cross section
a exponent
The values of the exponent a are taken from a UK proposal based on [5]. The exponent has been
slightly adjusted according to [6]. These values can be used in the absence of more accurate
values.
C5.9. Lateral instability of slender beams
Compared to the ENV, the following changes have been made:
1. It is clearly stated that the check of lateral instability of beams is relevant in situations where
lateral bracing is lacking. For beams in finished structures, lateral instability is normally
prevented by lateral bracing from adjacent members (e.g. floor or roof elements).
2. A lateral deflection l/300 has been introduced as an imperfection to be used in calculations
concerning lateral instability and balance at supports.
3. The criterion for neglecting second order effects is different. It is explained and compared to the
ENV below.
Expression (5.40) is based on a numerical study [7]. It is technically equivalent to the corresponding
criterion in the new DIN 1045 [8], but it has a different mathematical formulation to show the main
parameters l/b and h/b more clearly.
Figure 5.36 shows a comparison according to [7] between the numerical results and expression
(5.40). The corresponding criterion according to the ENV is also shown. It is quite clear that the
ENV criterion does not represent the numerical results very well; it is too conservative in many
cases and unsafe in other cases. The DIN model is much better.
Figure 5.36. Criteria for ignoring second order effects in beams according to ENV and EN
in comparison with numerical results [7]
In the final version, a distinction between persistent and transient design situations has been
introduced, together with an additional criterion for h/b; this is based on national comments.
5.10 Prestressed members and structures
See example 6.15
5.11 Analysis for some particular structural members
No specific comment on this part.
Appendix 1. Verification of new model for slenderness limit
Variables covered:
Concrete grade: C20, C40, C80
Guide to EC2 Section 5
Page 535 Table of contents
Reinforcement ratio: e = 0,1 and 0,5 (for C40 also e = 0,3)
Effective creep ratio: ¢ef = 0 and 2 (for C40 and e = 0,3 also ¢ef = 1)
Explanations to diagrams:
Horizontal axis: relative normal force n
Vertical axis: slenderness limit Ȝlim
Full curves: 10%criterion, alternative 1
Dashed curves: 10%criterion, alternative 2
Thick grey curves: new proposal for slenderness limit
Concrete C20
Concrete C40
Guide to EC2 Section 5
Page 536 Table of contents
Concrete C80
Appendix 2. Calibration of simplified methods
A2.1 Main calculations and results
Calculations have been made for isolated columns with the following variables:
The total number of individual cases is 4ā 9ā 3ā 3ā 3 = 972 (not including ì = 0 and e0 = 0).
For each case, calculations have been made with the general method according to 5.8.6 and with
the simplified methods, i.e. the stiffness and curvature methods according to 5.8.7 and 5.8.8
respectively. The results are summarized in table A21 on the following two pages.
The vertical axes in the diagrams represent the ratio
Maximumfirst order moment according to simplified method
Maximumfirst order moment according togeneral method
The moment ratio is given with the mean value m and with m+s and ms respectively, where s is
standard deviation. The mean value (and standard deviation) for a certain value of the independent
variable includes all the values for the other variables. Thus, for one value of e.g. the slenderness,
the mean value and standard deviation of the moment ratio represent 972/4 = 243 individual values;
for one value of the eccentricity 972/9 = 108 values etc.
The horizontal axes represent the main variables: slenderness ì, eccentricity e0/h, reinforcement
ratio (both e and ȡ), concrete strength fck and effective creep ratio ¢ef.
Interaction diagrams have also been prepared, covering all the abovementioned cases and
including the different methods and alternatives. However, to present all these diagrams would
require too much space, and it would be difficult to obtain an overall view of the results.
Therefore, only two such diagrams will be shown to illustrate certain aspects.
12
In [1] the simplified methods are compared with the general method also for columns with circular cross section and
with a different distribution of the first order moment. Although [1] deals with more refined versions of the stiffness and
curvature methods, the main conclusion is that the same models, as for rectangular section and constant moment, can
be used also for other cross sections and variations of the first order moment.
Guide to EC2 Section 5
Page 537 Table of contents
Table 5.A21. Summary of comparisons between simplified methods and general method
Table 5.A21, continued
Guide to EC2 Section 5
Page 538 Table of contents
A2.2 Discussion
All the simplified methods show a rather wide scatter when compared to the general method.
This is inevitable; a method giving close agreement with the general method over a wide range of
parameter values would no longer be simple.
An illustration is given in figure A21. The “curvature method” gives reasonable results for low to
moderate slenderness, but becomes extremely conservative for high slenderness ratios.
This is because the factor Kr (K2 in ENV) gives no reduction of the curvature at all when n < 0,4, and
for high Ȝ values n is practically always < 0,4. The same is true for the simplest version of the
stiffness method (expr. 5.26). With correction of the stiffness for normal force and slenderness,
expression (5.22) to (5.24), the result is much improved.
It is difficult to calibrate a simple method so that it accurately follows the general method,
particularly for high slenderness ratios and small eccentricities. This is true for all methods, see the
first and second rows of diagrams in table A21.
In figure A21 there is no effect of creep (¢ef = 0). Figure A22 shows the corresponding curves for
¢ef = 2 (a comparatively high value). Two sets of curves are given for the curvature method. The
upper curves are based on K
¢
= 1, corresponding to the method in ENV 19921 1, 4.3.5.6.3, where
there is no effect of creep. The lower curves are based on K
¢
according to expression (5.37).
Without effect of creep (= ENV), the curvature method is consistently unsafe for low and moderate
slenderness. This can be seen also in the third column of diagrams in table A21.
With K
¢
according to expression (5.37), creep is well taken into account.
Guide to EC2 Section 5
Page 539 Table of contents
References
[1] Westerberg, B: Design Methods for Slender Concrete Columns. Tyréns Technical Report
1997:1. Stockholm, September 1997.
[2] FIP Recommendations, Practical Design of Structural Concrete. fib (CEBFIP), September
1999.
[3] Design Handbook for High Performance Concrete Structures. Handbook published in
Sweden, 1999.
[4] Beeby, A W and Narayanan, R S: Designers’ Handbook to Eurocode 2, Part 1.1. Thomas
Telford, London, 1995.
[5] Bresler, B: Design Criteria for Reinforced Columns under Axial Load and Biaxial Bending.
ACI Journal, November 1960.
[6] Whittle, R T and Lawson, R: Biaxial bending with axial compression. An investigation into the
use of Bresler coefficients for determining the capacity of reinforced concrete sections under
combined axial compression and biaxial bending. March 2000
[7] König, G and Pauli, W: Nachweis der Kippstabilität von schlanken Fertigteilträgern aus
Stahlbeton und Spannbeton. Beton und Stahlbetonbau 87 (1992)
[8] Hellesland, J: On column slenderness limits. Mechanics Division, University of Oslo, 1999
0528
Guide to EC2 Section 5
Page 540 Table of contents
Guide to EC2 Section 6
Page 61 Table of contents
SECTION 6 ULTIMATE LIMIT
STATES (ULS)
6.1 Bending with or without
axial force
SECTION 6 ULTIMATE LIMIT STATES (ULS)
C6.1 Bending with or without axial force
6.1.1 Determining the compression resultant and its position compared to the edge of
maximum deformation in case of rectangular section
Two cases should be distinguished:
e) real neutral axis (x h)
f) virtual neutral axis (x > h)
Real neutral axis
a1) Diagram parabola – exponential – rectangle
The resultant C of the block of compressive forces related to a rectangle of width b and depth x is
expressed by
C = ȕ1 · fcd · b · x
and its position, measured starting from the edge where the strain is İcu2, is defined by ȕ2 x.
The formulae of ȕ1 and ȕ2 , in function of strain İc , are:
c
o c
 c =
c
)
cu 2
c
0
1 cu2
cd cu2
d
( )
f
c
c
o c c c
 c = ÷
c o c c
)
)
cu 2
cu 2
c
0
2 cu2
cu2 c
0
( ) d
( ) 1
. ( ) d
The numeric values of ȕ1 and ȕ2 are shown in function of fck in Table 6.1 .
In all tables limit the number of decimals to 3 maximum e.g. 0,80952 = 0,810 etc.
Table 6.1.Values of ȕ1 and ȕ2
a2) Rectangular diagram
With the combined effect of the Ȝ and Ș factors recalled in Chapt. 3.1.7, the following values result:
ȕ1 = ì · Ș
ȕ2 = ì /2
the values are shown in Table 6.2.
Table 6.2. Values of ȕ1 and ȕ2 for rectangular diagram
b) Virtual neutral axis
b1) Parabola – exponential – rectangle diagram
With reference to Fig. 6.1, where
ç' = x/h
the maximum and minimum strain at the section, respectively İt and İb (top and bottom) are given by
the formulae
ç c
c =
c
+ ç ÷
c
c2
t
c2
cu2
'
' 1
 
c = ÷ c

ç
\ .
b t
1
1
'
fck (N/mm
2
) up to 50 55 60 70 80 90
Ƣ1 0,80000 0,76781 0,73625 0,67500 0,61625 0,56000
Ƣ2 0,40000 0,39375 0,38750 0,37500 0,36250 0,35000
fck (N/mm
2
) up to 50 55 60 70 80 90
Ƣ1 0,80952 0,74194 0,69496 0,63719 0,59936 0,58333
Ƣ2 0,41597 0,39191 0,37723 0,36201 0,35482 0,35294
Guide to EC2 Section 6
Page 62 Table of contents
Figure 6.1. Rectangular section with virtual neutral axis
Indicating respectively by ȕ1t and ȕ2t the resultant and its position about the whole length x and by ȕ1b
and ȕ2b the similar quantities about the xh part, the resultant ȕ3 and its position ȕ4 compared to the
most compressed edge and in relation with depth h are given by:
 = ç  ÷ ç ÷ 
3 1t 1b
' ( ' 1)
ç   ÷ ç ÷  ç ÷  +
 =

2
1t 2t 1b 2b
4
3
' ( ' 1) (( ' 1) 1)
The ȕ3 and ȕ4 values for a number of x/h ratios are given in Table 6.3.
Table 6.3. ȕ3 and ȕ4 values
b2) Rectangular diagram
In this case (x h) EC2 does not give instructions. It is nevertheless possible to write a formula that
gives the equivalent depth h* in relation with x, so that it results
*
x  h
h = h
x  kh
ì
where the k a factor is
determined by imposing that for x = h is h* = ì · h. It results:
1
k = 2 
ì
so that the equivalent depth h* is obtained from the expression:
*
x  h
h = h
1
x  (2  ) h
ì
ì
Values of Ȝ, Ș, k are given in function of fck in Table 6.4.
Table 6.4. Values of ì, Ș, k
As an application, the values of ȕ3 and of ȕ4 were calculated in analogy to that was developed for
case b1). The resulting values are given in Table 6.5.
fck (N/mm
2
) ƫ Ƨ k
s 50 0,80000 1,00000 0,75000
55 0,78750 0,97500 0,73016
60 0,77500 0,95000 0,70968
70 0,75000 0,90000 0,66667
80 0,72500 0,85000 0,62069
90 0,70000 0,80000 0,57143
parabola – rectangle costitutive law
h
x
fck = 50 N/mm
2
fck = 55 N/mm
2
fck = 60 N/mm
2
fck = 70 N/mm
2
fck = 80 N/mm
2
fck = 90 N/mm
2
Ƣ3 Ƣ4 Ƣ3 Ƣ4 Ƣ3 Ƣ4 Ƣ3 Ƣ4 Ƣ3 Ƣ4 Ƣ3 Ƣ4
1,00 0,80952 0,41597 0,74194 0,39191 0,69496 0,37723 0,63719 0,36201 0,59936 0,35482 0,58333 0,35294
1,20 0,89549 0,45832 0,83288 0,43765 0,78714 0,42436 0,72968 0,41022 0,69249 0,40355 0,67720 0,40186
1,40 0,93409 0,47480 0,88197 0,45841 0,84129 0,44724 0,78831 0,43492 0,75381 0,42907 0,73986 0,42761
1,60 0,95468 0,48304 0,91168 0,46990 0,87615 0,46046 0,82826 0,44975 0,79679 0,44461 0,78422 0,44335
1,80 0,96693 0,48779 0,93113 0,47702 0,90007 0,46895 0,85695 0,45954 0,82834 0,45499 0,81702 0,45389
2,00 0,97481 0,49077 0,94460 0,48178 0,91730 0,47478 0,87838 0,46644 0,85234 0,46237 0,84211 0,46140
2,50 0,98550 0,49475 0,96464 0,48861 0,94420 0,48347 0,91348 0,47705 0,89255 0,47385 0,88448 0,47311
5,00 0,99702 0,49893 0,99060 0,49705 0,98285 0,49512 0,96937 0,49234 0,95972 0,49089 0,95622 0,49057
Guide to EC2 Section 6
Page 63 Table of contents
Table 6.5. ȕ3 and ȕ4 values for rectangular diagram
6.1.2 Calculation of strength of rectangular section
6.1.2.1 Determination of NRd and MRd
Given a transverse rectangular section with symmetrical geometry and reinfocement, the reinforcing
bars, the materials and the line that defines the deformed configuration at ultimate limit states, the
design normal force and the design bending moment are determined about the centroidal axis.
Figure 6.2. Rectangular section at ultimate limit state
On the hypothesis that straight sections remain straight, deformation are as in fig. 6.2. On the basis of
their level from the stressstrain diagrams of concrete and steel, the corresponding stresses are
calculated.
In order to determine NRd and MRd two equations of equilibrium (horizontal shift and rotation) are
written.
Equilibrium to shift: if Nc is the resultant of compressive stresses applied to concrete, N’s the resultant
of stresses applied to the compressed reinforcing bars A’s and Ns the resultant of traction in the
reinforcing bars As ,
NRd = Nc + N’s  Ns
The single terms can be developed as:
Nc =bā1āxāfcd
N’s = o'sāA’s
Ns = osāAs
In particular
o's = c'sāEs if
yd
s
s
f
<
E
' c
where
 
c = c ÷

\ .
s cu
d'
' 1
x
(6.1)
o's = fyd if
yd
s
s
f
E
' c >
likewise
os = csāEs if
yd
s
s
f
<
E
c
where
 
c = c ÷

\ .
s cu
d
1
x
(6.2)
os = fyd if
yd
s
s
f
E
c >
The design bending moment about the centroidal axis is expressed as
Rectangular costitutive law
h
x
fck = 50 N/mm
2
fck = 55 N/mm
2
fck = 60 N/mm
2
fck = 70 N/mm
2
fck = 80 N/mm
2
fck = 90 N/mm
2
Ƣ3 Ƣ4 Ƣ3 Ƣ4 Ƣ3 Ƣ4 Ƣ3 Ƣ4 Ƣ3 Ƣ4 Ƣ3 Ƣ4
1,00 0,80000 0,40000 0,76781 0,39375 0,73625 0,38750 0,67500 0,37500 0,61625 0,36250 0,56000 0,35000
1,20 0,88889 0,44444 0,85601 0,43898 0,82344 0,43339 0,75938 0,42188 0,69695 0,40997 0,63636 0,39773
1,40 0,92308 0,46154 0,89154 0,45720 0,86011 0,45269 0,79773 0,44318 0,73623 0,43308 0,67586 0,42241
1,60 0,94118 0,47059 0,91073 0,46704 0,88030 0,46332 0,81964 0,45536 0,75946 0,44674 0,70000 0,43750
1,80 0,95238 0,47619 0,92274 0,47320 0,89308 0,47004 0,83382 0,46324 0,77482 0,45577 0,71628 0,44767
2,00 0,96000 0,48000 0,93097 0,47742 0,90191 0,47469 0,84375 0,46875 0,78572 0,46219 0,72800 0,45500
2,50 0,97143 0,48571 0,94341 0,48380 0,91534 0,48176 0,85909 0,47727 0,80282 0,47225 0,74667 0,46667
5,00 0,98824 0,49412 0,96191 0,49329 0,93554 0,49239 0,88269 0,49038 0,82975 0,48809 0,77677 0,48548
Guide to EC2 Section 6
Page 64 Table of contents
= ÷ + o ÷ + o ÷
'
Rd c 2 s s s s
h h h
M N ( x) A' ( d') A ( d')
2 2 2
.
In case both the reinforcing bars are yielded (os = o's = fyd ),:
NRd = bāȕ1āxāfcd  A'sāfyd + Asāfyd
=  ÷ + ÷ ÷
Rd 1 cd 2 s s yd
h
M xbf ( x) ( A A' )f (d d')
2
(6.3)
ȕ1 and ȕ2 are factors given in Tables 6.1 or 6.2.
6.1.3. Design of reinforcing bars in case of bending without axial force and in case of bending
with great eccentricity axial force
Let's take a transversal section with axis of symmetry y (Fig. 6.3) and load effects in the plane of
symmetry. Given the design load effects at ultimate limit state MEd and NEd, the bending moment
about the tension reinforcement is calculated:
MEsd = MEd  NEd · ys
and the reinforcement area As required by MEsd is calculated like in the case of simple bending
moment. The axial force NEd is taken into account subsequently, by correcting of an equivalent
quantity the resistance of the tensioned reinforcement steel. In such cases, the applied axial force
must be taken into account with its sign: if NEd is a compression force, it will reduce the area of
tensioned steel.
Figure 6.3. Simple bending moment and composite bending moment with great eccentricity
A general rule is adopted: only the tensioned steel (As ) is provided, and only in case the tensioned
steel is not sufficient, some compressed steel (A's ) is added. In order to ensure that the structure has
a ductile behaviour, the strain İs of the tensioned steel must be greater that the strain corresponding
to the limit of elasticity, that is İs İyd = fyd/Es . This implies that the neutral axis does not exceed the
depth
c
=
c + c
cu
lim
cu yd
x d
where İcu is the strain at the compressed end. This limitations is valid for isostatic members; for other
cases, other limitations apply (see note).
The value of İcu depends exclusively on the concrete class. The İcu2 (for parabola and exponential –
rectangle diagram) and İcu3 (for bilateral and uniform) values are identical [Table 3.1EC2]. The İyd
value depends on the steel design stress fyd = fyk/¸s .
If MEsd is greater than the moment Mlim, that corresponds to xlim in presence of tensioned reinforcement
only, a certain amount of compressed steel has to be put in place. The difference
ǻMEsd = MEsd  Mlim
is to be absorbed by two sets of reinforcement, one compressed and one tensioned, of area
' Esd
s '
yd
ǻM
A =
f (d  d )
, which both work at the design limit of elasticity.
The area of reinforcement steel, A’s , has to be added to the section As corresponding to the limit
bending moment.
Note
The procedure exposed in the general guidelines follows from the principle of committing, as far as
possible, compression to concrete and traction to steel.
More severe limitations of x than those above are required in order to meet ductility requirements in
the case of indeterminate structures in bending, where redistribution of moments may take place.
Guide to EC2 Section 6
Page 65 Table of contents
Moreover, both in statically indeterminate and determinate cases, the verification of serviceability
tensional stresses [7.2(2)EC2] implicitly requires that the depth of neutral axis at ultimate limit state is
limited. In the design process point [9.2.1.1(1EC2] must also be taken into account requiring a
minimal quantity of tensioned reinforcement steel to avoid that fragility situations arise side steel. In
other words, it's necessary that the resisting bending moment of the reinforced section is greater than
the moment that causes cracking.
6.1.4 Rectangular section
6.1.4.1 Use of parabolarectangle and exponentialrectangle stressstrain relations
Design is simple if the ȕ1 and ȕ2 values (respectively the resultant and its distance from the edge, for
an element of unitary width and depth), given at point 7.2, are used.
Given the section dimensions (b, h, d, effective depth defined as distance of the tensioned
reinforcement steel centroid from the compressed edge), materials and action effects, the bending
moment (Fig. 6.3) about the tensioned reinforcement elements is calculated:
MEsd = MEd – NEd .ys .
In order to determine if As is sufficient, or if also A's is necessary, the following procedure is followed:
xlim is determined
the limit bending moment MRd,lim with tensioned reinforcement only is calculated
MRd,lim = Fc·zlim
where Fc = ȕ1·b·xlim·fcd is the resultant of compression stresses and zlim = (d ȕ2 ·xlim) is the inner lever
arm.
a) If MEsd is smaller than Mrd,lim, As alone is needed. In order to determine it the value of x
corresponding to MEsd must be defined.
It results:
ȕ1·b·x ·fcd· (d ȕ2 ·x) = MEsd
which, developed, becomes:
÷ + =
  
2 Esd
2 1 2 cd
M d
x x 0
b f
Solving:
 
= ÷ ÷

   
\ .
2
Esd
2 2 1 2 cd
M d d
x
2 2 b f
Remembering that MEsd = As fyd z, with z = (d ȕ2 ·x), finally
=
Esd
s
yd
M
A
f z
As Fc must be equal and contrary to Ft , the resultant of traction of the reinforcement steel As , it can
also be determined by:
As = ȕ1·b·x ·fcd/fyd
If MEsd is greater than MRd,lim , some reinforcement steel A's in compression is needed. To
calculate it, ǻMEsd = MEsd  MRd,lim. from which:
Esd
s
yd
M
A'
f (d d')
A
=
÷
The tensioned reinforcement is:
 
= + +

\ .
Rd,lim
s Ed s
yd lim
M 1
A N A'
f z
In such cases ǻMEsd must be sensibly smaller than MRd,lim , viz. it must not distort the problem.
Guide to EC2 Section 6
Page 66 Table of contents
6.1.4.2 Rectangular diagram of concrete stresses
With reference to Fig. 6.4, y = ì·x is the depth of the compressed zone and Ș·fcd is the design tensile
stress. Values of the ì and q factors are given in Table 6.4. It results:
Fc= b·y· Ș·fcd
z = (d – y/2)
MRd = Fc ·z = b·y· Ș·fcd ·(d  y/2)
Figure 6.4. Rectangular section with rectangular concrete stress diagram
Rearranging:
   
÷ + =
 
q \ . \ .
2
Esd
2
cd
2M y y
2 0
d d b.d ( f )
with the solution
= ÷ ÷
q
Esd
2
cd
2 M y
1 1
d b d ( f )
Also in this case, the considerations and developments of the previous point apply.
6.1.4.3 Tsections
Two situations can arise in Tsections:
the neutral axis is in the flange: no difference as if the section was rectangular;
the neutral axis crosses the web: its determination and the subsequent developments are simple
if the rectangular stress diagram provided for at paragraph [3.1.7(3)EC2] is adopted. An
approximate method is also presented.
6.1.4.3.1 General method
Introduction
As very high values of resisting moments can be reached with Tsections, especially if medium/high
strength concrete is used, the tensioned reinforcement steel should be laid on two layers, each one
of area As. The upper layer will have distance d from the compressed edge so that, as this
reinforcement layer will have at least strain İyd , the lower layer is also surely yielded. In some cases,
a lower bulb for the placement of tensioned reinforcement steel could be needed.
The rectangular stress diagram is adopted for calculation. The block of compressive stresses
is defined by a uniform value Șfcd and by the extension y = ìx where ì and Ș are two factors
lower than 1 and function of fck according to the formulae [3.19 to 3.22EC2]. Values of and Ș
are given in Table 2.3.
In case of simple bending moment, the equilibrium to rotation between external and internal
moment is written with reference to the layer that corresponds to half way between the
reinforcement layers As (steel reinforcement centroid). In this way the contributions of the two
tensioned reinforcement layers do not appear in the equation. The position of the neutral axis
is determined through this equation. Then, the reinforcement elements are determined by
equilibrium to shifting.
In case of bending with axial force the reference layer of bending moments M* (sum of the
given moment and of the one deriving from the shifting of the force NEd) will still be the above
indicated one. In this case the equilibrium to shifting, that is used to determine As, also
contain NEd.
Procedure
With reference to Fig. 6.5 y is assumed as the basic parameter. M* is the given bending moment
about the As reinforcement centroid. s is the distance of this centroid from the concrete compressed
edge. The bending moment resistance is expressed as the sum of the moments of blocks of
compression, about the same layer. The equilibrium gives:
Guide to EC2 Section 6
Page 67 Table of contents
( )
cd w w
y c
Șf b y s  + b  b c s  = M*
2 2
(    
 
(
\ . \ . ¸ ¸
By developing, it results:
( )
( )
2
w
w cd
2M* c
y  2sy + b  b c s  = 0
b Șf 2
÷
 

\ .
Figure 6.5. Tsection treated with the rectangular stress diagram
Taking:
( )
w
w
2 c
k = b  b c s 
b 2
 

\ .
with k function of the geometrical data only, y is given by:
( )
2
w cd
2M*
y = s  s  + k
b Șf
If M* is lower than the limit moment, no compressed reinforcement bars are needed. The tensioned
reinforcement area is given by:
2As = (Fc +NEd )/fyd
and the resultant of compression Fc is:
Fc = q fcd·[bw·y + (b – bw)·c]
If it is not necessary to put the tensioned reinforcement bars on two layers, in the aboveshown
formulae is is sufficient to identify s as d, while the first member of (7.13) gives the necessary area of
reinforcement.
6.1.4.3.2 Approximate design method
The method applies to those Tbeams where the flange is able to withstand all compressive forces
deriving from bending moment and axial force, without addition of compressed reinforcement bars. It's
assumed that tensile stresses are uniformly distributed.
With reference to Fig. 6.6, the total bending moment M* about the centroid of tensioned reinforcement
bars is calculated: M* = MEd  y*· NEd (NEd positive if traction).
Fig.6.6. Tsection. Approximate design method
The reinforcement is given by the formula
*
s Ed
yd
1 M
2A = + N
c f
s 
2
(
(
(
 
(

(
\ . ¸ ¸
and it must be verified that the average compressive stress in the flange is not greater than fcd:
*
c cd
M
ı = f
c
b c s 
2
s
 

\ .
Guide to EC2 Section 6
Page 68 Table of contents
6.1.5 Bending with axial force in rectangular section with symmetric reinforcement bars
The ultimate limit state behaviour of a rectangular section with reinforcement bars As = A's placed
symmetrically at the section edges and subjected to simple bending moment with an eccentric axial
force NEd is studied. The analysis is carried out through the determination of the moment resistance
MRd about the centroidal axis for given values of NEd and geometric and mechanical properties of the
section.
The study starts from Fig. 6.1 which is reproduced as far as it's needed in Fig. 6.7. Four deformed
configurations (1234) are taken into account. The first three, obtained by rotation around point B of
the straight line that represents the plane section, are characterized by particular depths of the neutral
axis x; line 4, that configures uniform strain İc2, is vertical in the representation and passes by point C.
Fig.6.7. Possible strain distributions at the ultimate limit state
Relations that are recalled in the treatment:.
Strain of upper reinforcement A's (see point 6.1.2.1):
 
c = c ÷

\ .
'
s cu2
d'
1
x
Strain of lower reinforcement As (see point 6.1.2.1):
 
c = c ÷

\ .
s cu2
d
1
x
Definition of line 1
Passing by B, the straight line assumes configuration 1 characterized, at the level of the upper
reinforcement layer, by strain
c = c = =
¸
yk '
s syd
s s
f
0, 00196
E
(design limit of elasticity for B450 steel
with tyk = 450 N/mm
2
and Es = 200 kN/mm
2
). The depth of the neutral axis, calculated by (6.1) results:
 
c
= = 

c ÷c
\ .
cu2
1 1
cu2 syd
x d' k d'
The k1 values for different concrete classes (as İcu2 is function of fck ), are given in Table 6.6
(İcu2 in absolute value).
For x < x1 the upper reinforcement layer is in elastic field; for x x1 it works at stress fyd. The lower
reinforcement layer, for x x1 , works at stress fyd .
Table 6.6 values of k1 and k2
Line 2: it is defined by the strain İs = İsyd of the lower layer of reinforcement. In this case the depth of
the neutral axis, deduced by (6.2) results:
fck
(N/mm
2
)
İcu2 k1 k2
s 50 0,0035 2,27 0,64
55 0,0031 2,71 0,61
60 0,0029 3,08 0,59
70 0,0027 3,65 0,58
80 0,0026 4,06 0,57
90 0,0026 4,06 0,57
Guide to EC2 Section 6
Page 69 Table of contents
 
c
= = 

c + c
\ .
cu2
2 2
cu2 syd
x d k d
k2 values are shown in Table 6.6.
For positions of the neutral axis x1 x x2, both reinforcement layers are subjected to a stress
ıs = fyd (compression for the upper one, traction for the lower one), with strain İs İsyd.
Configuration 3 is characterized by strain value İs = 0 (and therefore ıs = 0) for the lower
reinforcement (x = d). Therefore in the 34 range stress is ıs = fyd for the upper reinforcement and
ıs = İs·Es for the lower.
Line 3' is defined by İc = 0 at the lower edge of section.
In configuration 4: İs = İs' = İc2 (in absolute value), which is always greater than İsyd. It results then
ıs = ıs' = fyd . In the transition from 3 to 4 the upper reinforcement is compressed at stress fyd, the
lower reinforcement at stress increasing from 0 to fyd.
The following values of the axial force resistance NRd correspond to configurations 1,2,3,4:
NRd1 = ȕ1bx1fcd
NRd2 = ȕ1bx2fcd
In these two cases there are no contributions from the reinforcement bars because these are
subjected to ±fyd, and generate two equal and opposite forces in equilibrium.
NRd3 = ȕ1bdfcd + A'sfyd
NRd4 = bhfcd +2Asfyd.
ȕ1 values are given in Table 6.1 from the parabolaexponential rectangle model and in Table 6.2 for
the rectangle model.
Calculation of the moment resistance in the four abovedefined sectors.
a) NEd < NRd1 , that is x < x1
The position x of the neutral axis must be preliminary determined by the equation of equilibrium to
shifting. Keeping in mind that the upper reinforcement A's is compressed in elastic field and that the
lower reinforcement As is tensioned at stress fyd, the equilibrium is written as:
s s 1 cd s yd Ed
ı' A  ȕ x b f + A f =  N
Taken ı's =Eİ's where İ's is given by (6.1), it results:
Ed s yd cu2 s s 2 cu2 s s
1 cd 1 cd
N + A f  İ E A
İ E A d'
x  x  = 0
ȕ b f ȕ b f
   
 
\ . \ .
The equation, written in synthesis
( ) ( )
2
x  * * x  * * * = 0
has the solution
2
1 1
x = + (**) + (**) + (* * *)
2 4
The ı's stress is now known and adds up to
'
s s cu2
d'
ı = E İ 1
x
 

\ .
and the moment resistance about the centroidal level is:
Rd s yd s s 1 cd 2
h h h
M = A f (  d') + A ı '(  d') + ȕ x b f (  ȕ x)
2 2 2
b) NRd1 NEd NRd2
As both reinforcements are yielded, NEd is exclusively supported by concrete. The equilibrium
equation is:
NEd = NRd = ȕ1b x fcd , so that the depth of the neutral axis is:
Ed
1 cd
N
x =
ȕ b f
Guide to EC2 Section 6
Page 610 Table of contents
With the abovedetermined x value, the moment resistance is:
( )
Rd s yd 1 cd 2
M = A f (d  d') + ȕ x b f 0, 5h ȕ x (6.4)
c) NRd2
s
NEd
s
NRd3
The equation of equilibrium to shifting is:
'
yd s 1 cd s s Ed
f A  ȕ x b f + A ı = N (6.5)
As the reinforcement As is in elastic field, ıs = Escs e cs is given by (6.2)
Replacing and developing it results:
'
Ed s yd cu2 s s 2 cu2 s s
1 cd 1 cd
N  A f  İ E A
İ E A d
x  x  = 0
ȕ b f ȕ b f
   
 

\ . \ .
(6.5bis)
Once determined x e ıs, the moment resistance results:
= ÷ + o ÷ + ÷
Rd s yd s s 1 cd 2
h h h
M A' f ( d') A ( d') xbf ( x)
2 2 2
( 6.6)
d) In the fourth field (NRd3
s
NEd
s
NRd4 ) the moment resistance can be determined, with a good
approximation, by the relation of proportionality indicated in fig. 6.8, which shows the final end of the
interaction diagram MN.
Figure 6.8. Terminal end of the interaction diagram MN
The moment resistance reaches a maximum for x = x2 where the analytic function that expresses it
has an edge point due to the discontinuity between (6.4) and (6.6). The derivative for x = x2 is positive
if (6.4) is used and is negative with (6.6).
6.1.6 Interaction diagram MRdNRd
In the case of sections subjected to bending with axial force with small eccentricity, such as those of
columns, the most logical solution is the one with double symmetric reinforcement. Such sections can
also withstand simple bending and, if it's the case, composed bending in relation if the dimensions
and placement of reinforcement.
In order to have an overview on the problem, let's consider the load capacity of a rectangular section
(Fig. 6.9) with dimension h = 600 mm, b = 400 mm of fck 30 concrete, in the four conditions:
no reinforcement
symmetric reinforcement (fyk = 450 N/mm
2
) at the edges in percentage 0,5 – 1,0 – 1,5 on
each edge, distanced at d' =50 mm.
The calculation of the resistance of this section at the ultimate limit state is developed, for each of the
four reinforcement conditions, by associating the parabolarectangle diagrams for concrete and the
bilinear diagram or steel, with 7 configurations characterized by strain values of plane section
(reference to Fig. 1) given in Table 6.7. Steel B450 (fyk = 450 N/mm
2
).
Table 6.7. Deformed section configurations
Development of the calculation in relation with the third strain condition for the reinforced section with
1% bilateral reinforcement (As = A's = 2400 mm
2
)
Upper reinforcement: İ's =  0,0023 and therefore ı's = 391 N/mm
2
; F's = 939 kN
Lower reinforcement: İs = + 0,010 and therefore ıs = + 391 N/mm
2
, Fs = +939 kN
Neutral axis: x/d = 3,5/(3,5+10) , and as d = 550 mm, x = 142,6 mm
İc =  0,0035 at upper end İs = + 0,05000 (bottom reinf.)
İc=  0,0035 at upper end İs = + 0,02500 (bottom reinf.)
İc=  0,0035 at upper end İs= + 0,01000 (bottom reinf.)
İc=  0,0035 at upper end İs= + 0,00196 (bottom reinf.)
İc=  0,0035 at upper end İs= 0,00000 (bottom reinf.)
İc=  0,0035 at upper end İc= 0,00000 at lower end
İc=  0,0020 everywhere İc =  0,0020 everywhere
Guide to EC2 Section 6
Page 611 Table of contents
Fc = ȕ1bxfcd = 0,8095400142,617,0 = 785 kN
Distance of Fc from the compressed edge: ȕ2x = 59 mm
NRd = Fc + Fs + F's = 785 – 939 + 939 = 785 kN
Moment about the concrete section centroid
MRd = Fc (d/2 ȕ2x ) + Fs (d – d’)= 189,1 + 469,5 = 658,6 kNm
The eccentricity of NRd is MRd / NRd = 839 mm.
Fig. 6.9 graphically shows the 29 pairs of results obtained. For each value of the percentage of
reinforcement the points are joined by a straight line. The result is convex polygons.
The following observations arise from the observation of Fig. 6.9:
3. the polygons include the domains of resistance: the points of coordinates MEd, NEd placed
inside the polygon are in a safe zone; the points on the polygon strictly verify the ultimate limit
state; external point do not meet resistance conditions at ultimate limit state
4. a straight line parallel to the N axis intersects the polygon in two points. This means that with
a given reinforcement, a given bending moment can be withstood by two different values of
the axial force. The limit case of a single N value happens when the straight line passes by
the highest point of the polygon.
5. Only one value of M can be associated to a given value of N, as it's possible to verify by
tracing a line parallel to the M axis.
6. The polygon related to the nonreinforced section denotes the possibility to withstand bending
moments only if they come with an adequate axial force (provided by self weight or by
prestressing).
7. The branches on the left of the M axis denote resistance to bending with positive axial force
8. The straight line of inclination M/N = h/30 = 20 mm defines the field of use of the polygons: in fact,
according to [6.1(4)EC2] a minimal eccentricity must always be taken into account, adding up to
the bigger value between h/30 and 20 mm.
Such polygons as those traced below are called “MN interaction diagrams”.
Figure 6.9. Interaction diagram for rectangular section
6.1.7 Biaxial bending and bending with axial force
Biaxial bending may be separated into separate uniaxial bending components under circumstances
laid down in Eurocode 2 5.8.9 . For pure biaxial bending, where the bending components lay on the
two centroid axis of inertia, the problem solution has computational difficulties. If a design software is
not used, and calculation developed by hand, it should be processed by iterations; in such case it's
convenient to adopt the rectangular stress diagram for concrete.
Given a section subjected to an axial force NEd applied on the centre of gravity, and to two bending
components MEyd e MEzd expressed by two vectors orientated along a couple of orthogonal axis y e z
with origin in the centre of gravity. In general, the existence of a stress distribution that gives place to
resistance greater or equal to the action effects must be demonstrated, as well as the fact that the
straight line that connects the centre of gravity of the compressed zone with the centre of gravity of
the tensioned reinforcement is perpendicular to the resultant bending vector MEd. This implies that the
eccentric axial force must also lay on that line.
Guide to EC2 Section 6
Page 612 Table of contents
6.2 Shear
6.2.2 Members not requiring
design shear reinforcement
C6.2.2 Shear capacity of members without shear reinforcement
6.2.2.1. Shear flexure capacity
Most shear failures occur in the region of the member cracked in flexure. It is necessary to make a
distinction between shear flexure and shear tension. In this chapter only shear flexure is regarded,
which can be considered as the general case.
In ENV 199211 the equation for the shear capacity of members without shear reinforcement was
VRd1 = [tRd k (1.2 + 40 ȡl) + 0.15ıcp]bwd (6.7)
Where
tRd basic shear strength, which follows from tRd = 0.25fctk,0.05/¸c.
k factor allowing for the size effect, equal to k = 1.6 – d (m) > 1
ȡl flexural tensile reinforcement ratio, As/bwd < 0.02
ıcp design axial stress (if any) = NSd/Ac
bw minimum web section
The are two shortcomings with regard to the use of this equation. At first the role of the concrete
strength is not correct, as was demonstrated in [Walraven, 1987, pp. 68  71.] For lower strength
concrete classes the deviations were not yet very large, but if the strength increases the deviations
soon reach an unacceptable level.
The second problem is that the equation has principally been derived for beams, failing in shear
flexure and is not valid for members which typically fail in shear tension. Such members are for
instance prestressed hollow core slabs, which nearly always fail in shear tension, in the area where
the member is not cracked in flexure. Applied to such members Eq. 6.7 would give unnecessary
conservative results.
The recommendations for the determination of the shear flexure capacity of members not reinforced
in shear are given in chapter 6.2.2 of prEN 199211:2001. The basic formula is given as Eq. 6.2.a in
this document. This equation has been derived in the following way. The basic equation adopted,
which was believed to take appropriate account of the most important influencing factors like concrete
strength, longitudinal reinforcement ratio and crosssectional height was
( )
1/3
u l c w
V = C k 100 f b d µ (6.8)
where
k = size factor = 1 + (200/d)
1/2
ȡl = longitudinal reinforcement ratio
fc = concrete cylinder strength (N/mm
2
)
C = coefficient to be determined
A selection was made of a representative number of shear tests, considering a parameter variation as
wide as possible and as well as possible distributed within practical limits. This was already done by
König and Fischer (1995). An overview of the test parameters is given in Fig. 6.10.
Then for every test result the optimum value C was determined. If the distribution is normal, Fig. 6.11
a lower bound value for C was determined according to the level 2 method described in [Taerwe,
1993 ] with the equation:
Clower bound = Cmeanā (1oā ā v) (6.9)
where
o sensitivity factor, equal to 0.8 for the case of one dominating variable (concrete strength)
ȕ reliability index, taken equal to 3.8 according to [Eurocode, Basis of Structural design, Draft
version 2001 ]
v standard deviation
If the distribution turns out to be lognormal, Fig. 6.11, the equation is
Clower bound = Cmeanā exp (ov ÷ 0,5v
2
) (6.10)
In these equations a reliability index ȕ = 3.8 means a probability of occurrence of 0.0072%.
König and Fisher (1995) carried out this procedure for 176 shear tests.
Guide to EC2 Section 6
Page 613 Table of contents
Figure 6.10. Relative frequency of parameters in test data bank used by König and Fischer (1995) in order to find a
reliable lower bound equation for the shear capacity of members without shear reinforcement
Figure 6.11. Normal and lognormal distribution
As a result of their analysis they found that a coefficient C = 0.12 would be a good lower bound. In
Fig. 6.12 it is shown that the prediction accuracy of this equation is substantially better than that of the
old EC199211 formula.
Figure 6.12. a. Shear capacity according to Eq. 6.8 (MC 90): relative frequency for NSC and HSC (König, Fischer,
1993)
b. Shear capacity according to Eq. 6.7 (ENV 199211): relative frequency for NSC and HSC, according to
König, Fischer (1993)
As an addition Fig. 6.13 shows an evaluation carried out by Regan [Regan, 1993], which confirms the
findings by König/Fischer.
Guide to EC2 Section 6
Page 614 Table of contents
Figure 6.13. Shear strength of nonprestressed members without shear reinforcement, comparison of test results with Eq.
6.8 [Regan, 1999.]
It was however argued, that the equation
VRd,c = 0,12 k (100 µl fck)
1/3
bw d (6.11)
has two disadvantages the first is that it does not distinguish between persistent & transient loading
combinations and accidental loading combinations, for which different safety levels apply (prEN 1992
11:2001 chapter 2.4.1.4 gives ¸c = 1,5 for persistent and transient and ¸c = 1,2 for accidental
situations). Therefore the equation was modified by introducing the concrete safety factor explicitly.
VRd,c = (0,18/¸c) k (100 µl fck)
1/3
bw d (6.12)
The second is that the shear capacity goes to 0 when ȡl = 0.
Furthermore it was wished to have a simple conservative value for VRd,c for a first check of the bearing
capacity. In many countries simple formulations have been used on the basis of
VRd,c = C fctd bw d (6.13)
where fctd is the design tensile strength of the concrete and C is a coefficient. Practice in the various
countries however is quite different because C varies in the range from 0,3 to 0,75.
Considering the value of C it should be noted that this equation is a simplification of the rigorous one.
To have general validity, even for rare but still possible cases, C should be based on the most
unfavourable combination of parameters. That means that the governing case is a slab with a large
crosssectional depth d and a low longitudinal reinforcement ratio.
In his paper “Basic facts concerning shear failure”, Kani (1966) showed that shear failures are unlikely
to occur for longitudinal reinforcement ratio’s smaller than 0,6%. However, his “shear valley” was
based on beams with a crosssectional effective depth of only d = 270 mm. For larger depths the
critical value of ȡ0 decreases. Therefore a number of shear failures reported in literature have been
selected with large d and small ȡ0 values, see Table 6.8.
Guide to EC2 Section 6
Page 615 Table of contents
Table 6.8. Determination of C on the basis of selected tests
The most unfavourable values for C are 0,34, found for Aster&Koch’s tests Nr.11 and 16, with d = 500
and 750 mm and ȡ0 = 0,46 and 0,42% respectively.
So, with some rounding off a value C = 0,35 would be appropriate for the simplified design equation.
In prEN 199211:2001 a value 0,40 is used. An argument might be that the utmost part of the
practical cases consists of slabs with smaller depths, subjected to uniform loading, where the
maximum shear force does not coincide with the maximum moment, and the reinforcement ratio’s are
small enough to ensure failure by bending. The seldom case of a slab spanning in one direction, with
a high crosssection, a critically low reinforcement ratio and a line load just at the most critical position
from the support would then have a slightly lower safety. On the other hand formula’s should always
be safe enough to take account of any possible (not likely) case, which would be an argument in
favour of the use of 0,35.
Some questions may be raised with regard to the definition of bw being “the smallest width of the
crosssection in the tensile area”. Tests on tapered crosssections showed that there is certainly an
influence of the definition of the web width, as shown in Fig. 6.14, left (tests by Leung, Chew and
Regan, 1976). Fig. 6.14, right, shows that a definition of bw as the average width of the beam would
be appropriate for this case.
Figure 6.14. Shear resistance of beams with tapered crosssection (Leung, Chew and Regan, 1976)
In a more recent publication (Regan, 2000) the author opts for a definition of bw = 2/3 bmin + 1/3 bmax,
but admits at the same time that the available evidence is rather scarce. A possible compromise could
be to define bw as the average width of the part of the crosssection in tension, with a maximum of
1,25 of the minimum width.
The equations in prEN 199111:2001 contain as well a term 0,15 ıcp regarding the influence of an
axial force on the shear capacity, for instance by prestressing. Basically the influence of prestressing
can be taken into account as proposed by Hedman & Losberg (1978). It was argued that, with regard
to the behaviour in shear, a prestressed beam can be regarded as a reinforced beam after the
decompression moment has been reached. On the basis of this argument the shear resistance was
formulated as
VRd,c = Vc + Vp
where Vc is the shear resistance of a similar nonprestressed beam and Vp is the contribution of the
prestressing force to the shear capacity, which can be formulated as Vp = M0/a, where M0 is the
decompression moment and a is the distance from the load to the support, Fig. 6.15.
Figure 6.15. Calculation of contribution Vp from prestressing to the shear resistance
according to Hedman and Losberg (1978)
However, this method works well for the evaluation of laboratory tests but is less suitable for real
Guide to EC2 Section 6
Page 616 Table of contents
members mostly subjected to uniformly distributed loading. A solution is to replace M0/a by
M0/(Mx/Vx), where Mx and Vx are the bending moment and the shear force in the section considered.
However, this would complicate the shear design because then Vp would be different in any cross
section.
Another disadvantage is that Vp would go to infinity in a moment inflexion point, where Mx = 0.
It can simply be derived that for a rectangular crosssection with a width b, a height h and an
eccentricity of the prestressing force ep, the contribution Vp to the shear resistance is
Vp = Fp (1/6 + ep/h) (a/h) (6.14)
Assuming d = 0,85h this would result in;
Vp = 1,18 Fp (1/6 + ep/h) (a/d) (6.15)
In most tests on shear critical beams the ratio ep/h is about 0,35. With a/d varying between 2,5 and
4,0, like in most shear tests, this would mean that Vp would vary between 0,15ıcpb d and 0,25ıcpb d.
Evaluating test results it is therefore not amazing that the coefficient 0,15 turns out to be a safe lower
bound in shear critical regions.
Nielsen (1990) compared the shear equation in ENV 199211 which gives about the same results as
Eq. 6.2a in prEN 199211:2001 for moderate concrete strengths, with 287 test results and found that
it was at the safe side.
The effect of longitudinal compression should, of course, not be mixed up with the effect of the cable
curvature, which exerts a favourable transverse load on the member. This effect, known as the load
balancing effect, is introduced as a load (load balancing principle).
For axial tension in prEN 199211 the same formula is used, with a different sign for 0,15ıcp, so that
an axial tensile force gives rise to a slight reduction of the shear capacity. It should be noted that in
continuous beams there is tension in both top and bottom and excessive curtailment at sections of
contra flexure may lead to diagonal cracking and shear failure in such a region. This was the main
cause of failure in an actual structure [Hognestadt and Elstner, 1957]. If a structural member is well
designed for axial tension the shear capacity of the members is hardly reduced.
This was for instance shown by Regan [Regan, 1971 and 1999] who carried out a systematic
investigation into the effect of an axial tensile force on the shear capacity of both members
unreinforced and reinforced for shear. Tests have been carried out according to the principle shown in
Fig. 6.16. Beams with a rectangular crosssection were provided with nibs, enabling the transmission
of an axial tensile force in the middle part. The axial tensile force varied between 0 and 130 kN. The
force could be applied in two ways: before subjecting the member to transverse loading, or in
proportion to the transverse loading. In both cases the shear capacity was hardly influenced, although
the member sometimes showed wide open cracks across the total cross section in the moment
inflexion region.
Figure 6.16. Results of tests on beams subjected to axial tension, bending and shear, and failing
in shear [Regan, 1999]
For similar arguments, reference is made to [Bhide and Collins, 1989]
Guide to EC2 Section 6
Page 617 Table of contents
6.2.2.2 Shear tension capacity
In special cases, like for instance when pretensioned strands are used in members with reduced web
widths, such as in prestressed hollow core slabs, shear tension failures can occur, Fig. 6.17.
Figure 6.17. Shear tension failure
Figure 6.18. Calculation of shear tension capacity with Mohr’s circle
In this case failure occurs due to the fact that the principal tensile stress in the web reaches the tensile
strength of the concrete in the region uncracked in flexure. The principal tensile strength in the web
calculated using Mohr’s circle, Fig. 6.18, is equal to
2 2
I N N
1 1
ı =  ı + + ı
2 4
 
t

\ .
(6.16)
substituting t = VRd,ct S/bwI and ıN = oI ıcp the code’s expression EC2, Eq. 6.3
( )
2
w
Rd,ct ctd I cp ctd
I b
V = f  ı f
S
o (6.17)
is obtained.
6.2.2.3. Loads near to supports
In 6.2.2 (5) the equation (5) or in prENV 199211:2001 the Equation 6.2.a, is extended with a factor
(2d/x) in order to cope with the increased shear capacity in the case of loads applied near to supports.
According to this formulation, at a distance 0.5d < x < 2d the shear capacity may be increased to
VRd,ct = 0,12 k (100 µl fck)
1/3
(2d/x) bw d. (6.18)
This may need some explanation, since it might be argued that loads near to supports may be treated
with the rules given in EC2, 2001 version, chapter 6.5 “Design of discontinuity regions with strut and
tie models”.
However, there are many arguments in favour of the formulation according to Eq. 6.18:
 According to the formulations for the strut and tie model the capacity of the concrete struts only
depends on the strength of the concrete, see e.g. fig. 6.19.
Consequently, the maximum capacity is a function of the concrete strength and the width of the
support area.
Figure 6.19. Bearing capacity of short member according to strut and tie model
with defined maximum concrete stress in the struts
It can easily be seen that this is a very simplified representation of reality, since the capacity of such
a member results to be independent of the slenderness ratio a/d, which is known to have a strong
influence. Furthermore short members are prone to significant size effects. It was shown [Lehwalter
and Walraven, 1994), that the size effect in short members is the same for short and slender
members, so that here also the factor k = 1 + ¥(200/d) applies.
Lehwalter carried out tests on short members with various sizes, a/d ratio’s and support widths, and
compared the equivalent maximum stress in the concrete struts, Fig. 6.20. The dotted plane is valid
for a maximum stress 0,6 fc. It is seen that for lower a/d ratio’s the capacity is considerably higher
than the one obtained with the strut and tie model. It is seen furthermore that the limit 0.55 fc as
Guide to EC2 Section 6
Page 618 Table of contents
defined in 6.4.5 5(P) for struts with transverse tension is appropriate for a/d < 2.0, members with
depths until 1 m and a support width up to about 0.25d.
For a number of practical members, like in the case of corbels and pile caps, it is important to reduce
the size as much as possible. A more accurate formulation than the strut and tie model is therefore
useful in those cases.
 Another case is shown in Fig. 6.21. It is a part of a foundation caisson in the Storebaelt bridge, with a
slab of about 1 meter and wall distances of about 5 m. A substantial part of the counterpressure of
the soil is transmitted directly to the walls, so that the governing shear load is small. Without a
provision like the one given in Eq. 10, unnecessary shear reinforcement would be required.
Figure 6.20. Maximum stress in concrete struts as calculated on the basis of test results
(Walraven, Lehwalter, 1989)
Figure 6.21. Foundation slab in Storebaelt caisson
By introducing the distance x and determining the shear capacity in every cross section, also
combinations of loads (like two concentrated loads, or a uniformly distributed load and a concentrated
load) can be handled.
An important question is whether the multiplication factor should be (3d/a, 2.5d/a or 2d/). Regan
[1998], on the basis of the analysis of many experiments, concluded that:
a. For simply supported beams subjected to concentrated loads a factor (2.5d/a) is appropriate. This
is confirmed in Fig. 6.22 on the basis of tests by Baldwin and Viest (1958), Clark (1951), De Cossio
and Siess (1960), Küng, Lehwalter (198 ), Matthey and Watstein (1963), Morrow and Viest (1957),
Regan (1971) and Rogowski and MacGregor (1983)
b. For continuous beams with concentrated loads even (3d/a) gives safe results.
c. For simply supported beams subjected to distributed loading only (2d/a) gives safe results. The
diagram in Fig. 6.23 is based on tests by Bernaert and Siess (1956), Leonhardt and Walther
(1961), Rüsch, Haugli, Mayer (1962) and Krefeld and Thurston (1966). The figure shows graphs of
Vtest/Vcalc plotted against l/d, with Vcalc computed assuming values of 2d/x and 2.5d/x. Most, but not
95% of the points lie above Vtest/Vcalc = 1 for 2d/x, but less than half do so with 2.5d/x. The results
which are on the unsafe side with 2d/x need not to be of too much concern as Krefeld and
Thurston’s work includes many beams with low concrete strengths.
Guide to EC2 Section 6
Page 619 Table of contents
Figure 6.22. Results of tests on simply supported beams without shear reinforcement subjected to concentrated loads
(Regan, 1998)
Figure 6.23. Results of tests on simply supported beams without shear reinforcement subjected to distributed loads
(Regan, 1998)
6.2.2.4. Prestressed members without shear reinforcement
Four failure modes may be envisaged under combination of shear and bending within the extremity
region for such elements:
1. Exceeding the tensile strength of concrete in regions uncracked in bending as described by
expression (6.4).
2. Exceeding the shear resistance given by expression (6.2a) in presence of bending moments
greater than the cracking bending moment.
3. Anchorage loss due to bending cracks within transmission length.
4. Snap back failure at cracking bending moment outside the transmission length, when the
acting shear corresponding to the cracking bending moment exceeds the bearing capacity
calculated with expression (6.2a).
The designer should verify all the four failure mechanisms described above with particular care to the
snap back behavior.
Guide to EC2 Section 6
Page 620 Table of contents
6.2.3 Members requiring
design shear reinforcement
C6.2.3 Members requiring shear reinforcement.
6.2.3.1 Non prestressed members
In ENV 199211 the Standard Method and the Variable Inclination Method were offered as design
alternatives. The Standard Method, in spite of its acceptable accuracy, is from a physical point of view
unsatisfactory, because the “concrete term” is purely empirical and hides the physical reality. This
reality is, that a redistribution of forces occurs in the webs of shear reinforced concrete beams,
resulting in strut inclinations smaller than 45°, Fig. 6.24.
Figure 6.24. Redistribution of forces in a shearloaded web by strut rotation
Because of a smaller strut inclination, a larger number of stirrups is activated, and the shear capacity
is increased. A result of the smaller strut inclination is, however, that the stresses in the concrete
struts are larger, so that an appropriate upper limit to the shear capacity has to be defined. This
method, with strut inclinations smaller than 45° may be assumed for design, and is known as the
“variable strut inclination” method. This approach is not only attractive because of its agreement with
the physical reality, but also because it is a simple equilibrium method, giving a transparent view of
the flow of forces in the structure.
In ENV 199211, in chapter 4.3.2.4.4, the designer is allowed to choose the strut inclination between
0,4 < cot u < 2,5
which means that ș may be chosen between 21,8° and 68,2°. The choice of the lowest value mostly
leads to the most economic design. In this case the compression struts are supposed to rotate from
an initial value of 45* to a lower value of about 22°. If the strut inclination is ș and the (vertical) shear
reinforcement yields, a shear force
u,3 yw
Asw
V = z f cotș
s
(6.19)
Is transmitted, Fig. 6.25a. If the shear reinforcement yields, the truss can, by rotation of the
compression struts to a lower inclination, activate more stirrups for the transmission of the shear force
and, as such, extend the zone of failure.
Due to strut rotation, the stress in the concrete struts increases. Consequently, rotation can only
continue until crushing of the concrete occurs. For an ultimate compression stress fc1 in the concrete
struts, the corresponding shear force is (Figure 6.25b).
w c1
u,2
b z f
V =
cot ș + tanș
(6.20)
where fc1 = vfc
In those equations
bw web width
z inner lever arm ~ 0,9d
s stirrup distance
fyw yield stress of stirrups
ș inclination of concrete struts
Asw crosssectional area of one stirrup
v effectiveness factor, taking account of the fact that the beam web, which is transversally in
tension, is not as well suited to resist the inclined compression as cylinders used to
determine fc.
Fig. 6.25. a. Ultimate capacity Vu,3 for yielding stirrups
b. Ultimate capacity Vu,2 for crushing of concrete struts
Guide to EC2 Section 6
Page 621 Table of contents
For v generally the expression
v = 0,6 (1 – fck/250) (6.21)
is used (CEBFIP Model Code 1990).
By equalling 6.19 and 6.20 the maximum possible shear force and the corresponding inclination ș are
found. Fig. 19 shows the development of Vu,2 and Vu,3 for decreasing ș.
Figure 6.26. Dependence of Vu,2 and Vu,3 on the strut inclination ș
The expression for the ultimate nominal shear strength vu and the corresponding value of ș are then
( )
u
c1
1
f
v
¢ ¢ (6.22)
And
tanș =
1
¢
¢
(6.23)
Where
( )
u
u
w
V
=
b 0,9d
v
(6.24)
sw yw
c1
ȡ f
=
f
¢ (6.25)
with
fc1 = vfc
and ȡw is the shear reinforcement ratio according to ȡw = Asw/bws.
Eq. 6.22 represents a circle in a Vu/fc1  ¢ coordinate system, Fig. 6.27. It is found that when ¢ runs
from 0 to 0.5, ș runs from 0 to 45°. For ¢ > 0.5, the value of ș is constant and equal to 45°. However,
it was assumed that cot ș = 2,5 is a limit. In graphical terms this is shown as the linear cutoff in
Fig. 6.27.
Figure 6.27. Graphical representation of Eq. 16, with cutoff for cot ș = 2,5
Measurements of the deformation of the web in shear loaded Ibeams show typically a behaviour as
shown in Fig. 6.28. The diagram shows lines numbered from 1 to 4.
Line 1 In the beginning of shear loading the beams is uncracked in shear so that the principal
strain direction is 45°.
Line 2 At the formation of inclined shear cracks the principal strain direction decreases
Line 3 After having reached the stabilized inclined crack pattern a new type of equilibrium is
obtained. The behaviour is elastic: the (constant) principal strain direction depends on the
“stiffness ratio'’ in the cracked state.
Line 4 When the stirrups start yielding the web searches for a new state of equilibrium. By rotating
down to a lower inclination the beam activates more stirrups to carry the load. In the mean
time the compressive stress in the concrete increases. When the crushing strength of the
struts is reached the beam fails in shear.
Guide to EC2 Section 6
Page 622 Table of contents
Figure 6.28. Rotation of the concrete struts as measured on the web of beams with shear reinforcement, schematically
represented (Walraven, 1995 and 1999)
Fig. 6.29a shows the strut rotation as measured in three beams with different shear reinforcement
ratio’s (L = low, M = medium and H = High). It can be seen that for M and H the beams fail before the
stirrups have yielded. Fig. 6.29b shows a similar diagram for a series of high strength concrete beams
(fc § 90 N/mm
2
).
Figure 6.29. a Strut rotation as measured in beams with normal strength concrete (Walraven, 1995)
b.Strut rotation as measured in beams made of high strength concrete (Walraven, 1999)
Fig. 6.30 shows a verification of the combination of Eq. 6.19 and Eq. 6.20 with the limit cot ș = 2.5
with test results from Sörensen (1974), Regan and RezaiJorabi (1987), Placas and Regan (1971),
Leonhardt and Walther (1961), Kahn and Regan (1971), Moayer and Regan (1971), Hamadi and
Regan (1980), Muhidin and Regan (1977), Levi and Marro (1993) and Walraven (1999).
Figure 6.30. Non prestressed beams with vertical stirrups – relationship between shear strength and stirrup reinforcement
Guide to EC2 Section 6
Page 623 Table of contents
6.2.3.2. Prestressed members with shear reinforcement
If the same rules are applied to prestressed members with shear reinforcement, like in ENV 199211
and MC’90, it can be seen that there is apparently an increase of both safety and scatter, Fig. 6.30.
The test used for this figure are from Hanson and Hulsbos (1964), Bennett and Debaiky (1974),
Moayer and Regan (1974), Levi and Marro (1993), Lyngberg (1976), Aparicio and Calavera (2000),
Görtz and Hegger (1999), Leonhardt, Koch and Rostasy (1974). Many of the test results are collected
in a databank, described in (Walraven, 1987).
Figure 6.31. Experimental results of shear tests on prestressed beams with shear reinforcement, in comparison with the
calculated results according to the variable strut inclination method (no special web crushing criterion for prestressed
concrete)
In ENV199211 the effect of prestressing on the upper limit of the shear capacity VRd,i is partially
neutral and partially negative. In 4.3.2.2 (4). The following statement is found:
“In the absence of more rigorous analysis, at no section in any element should the design shear force
exceed VRd,2. Where the member is subjected to an applied axial compression, VRd2 should be in
accordance with the following equation:
VRd,red = 1.67 VRd,2 (1  ıcp.eff/fcd) < VRd,2 (6.26)
Fig. 6.32 shows the dependence of the upper limit for the shear capacity on the level of prestressing.
Fiugure 6.32. Reduction of maximum shear capacity by axial compressive stress according to
ENV 199211, Clause 4.3.2.2 (4)
Guide to EC2 Section 6
Page 624 Table of contents
In his “Commentaries on Shear and Torsion”, Nielsen (1990) states that prestressing has a positive
influence on the shear capacity of beams with shear reinforcement. He proposed to multiply v (which
was then equal to v = 0.7 – fck/200 > 0.5) with a factor
oc = 1 + 2.0 ıcp,eff / fc with ıcp,eff / fc < 0,5 (6.27)
A comparison with 93 tests results shows, however, that the expression is not sufficiently
conservative, to serve as a safe lower bound over the whole region of test results, Fig. 6.33. It should
furthermore be noted that all test results, used in the comparison, have ratio’s of ıcp/fc lower than 0,4
and that the multiplication factor is obviously applicable only for this region.
Another proposal for taking the influence of prestressing into account in the vvalue was given by
Fouré (2000):
 for small compression, with 0 < ıcp < 0,4fcd
oc = (1 – 0,67 ıcp/fctm) (6.28)
 for large compression, with 0,4fcd < ıcp < fcd
Figure 6.33. Comparison of results of shear tests with variable inclination truss analogy with effectivity factor v = (0,7 –
fc/200)(1 + ıcp/fc), according to Nielsen (1990)
oc = {1,2 (1  ıcp/fcd) (1 + ıcp/fcd)}
0,5
(6.29)
However, for the region of low compressive stresses this expression gives about the same results as
Eq. 6.27
A more moderate expression, taking into account the influence of prestressing, is
oc = (1 + ıcp/fc) for 0 < ıcp/fc < 0.25 fc
oc = 1.25 for 0.25fc < ıcp < 0.5fc
oc = 2.5 (1  ıcp/fc) for 0.5fc < ıcp < 1.0fc
In Fig. 6.35 the same data as used in Fig. 6.31 are evaluated using Eq. 6.29. It appears that the
safety margin and the scatter are reduced (the details of the calculation are found in Appendix 1). The
new proposal is compared with the other ones in Fig. 6.34.
Figure 6.34. Comparison of new proposal (Eq. 6.29) with original formulation in ENV 199211 (without influence of
prestressing) and proposals by Nielsen (1990) and Fouré (2000) for fck
Guide to EC2 Section 6
Page 625 Table of contents
6.3 Torsion
Figure 6.35. Experimental results of shear tests on prestressed beams with shear reinforcement, in comparison with the
calculated results according to the variable strut inclination method, with extension according to Eq. 6.29
6.2.3.3. Members reinforced in shear with loads near to supports
Similar to members without shear reinforcement, in members with shear reinforcement the load
bearing capacity is increased for loads near to supports. In the Standard Method, as formulated in
ENV 199211 this was taken into account by multiplying the “concrete term”, with a factor (2d/x).
However, the Variable Inclination Method not containing a concrete term, it was introduced here for
x/d < 2. The formulation for this case is then
VRd = VRd,ct + Aswā fywd sino (6.30)
The transmission of forces occurs according to Figure6.36.
Figure 6.36. Combination of truss and strut and tie
Measurements on shear reinforcement showed that the stirrups just adjacent to the load and support
area do not reach the yield stress, (Asin, 2000). Therefore the shear reinforcement is considered to be
effective only within the central 0,75 area between load and support.
See example 6.6 and 6.7
Guide to EC2 Section 6
Page 626 Table of contents
6.4 Punching C6.4. Basic equation for symmetrical punching at interior columns
6.4.1. Punching shear capacity of nonprestressed slabs without punching reinforcement
In ENV 199211 a nominal shear stress was defined as the load divided by the product the slab's
effective depth and the length of the control perimeter, which was defined to located at (1.5d) from the
edge of the column. The design punching shear capacity was
VRd1 = vRd1·u (6.31a)
Where u is the length of the critical perimeter, taken at a distance 1.5d from the loaded area, the
design shear resistance per unit length vRd1 followed from
vRd1 = tRd k (1.2 + 40ȡl)d (6.31b)
where tRd basic shear strength = 0.25 fctk /¸c
k size factor = 1.6 d [m] > 1.0
d effective depth of slab = (dx + dy)/2
ȡl flexural reinforcement ratio = (ȡx + ȡy)/2
It was already shown in 1980 [1], that in the derivation of this equation an error was committed, which
leads to unconservative results for higher concrete strengths. Therefore it was decided adopt the
formulation for the punching shear capacity given in Model Code 1990. when the design punching
shear capacity is given by
VRdc = vRdc .ud (6.32a)
where
u = length of the critical perimeter, taken at a 2d distance from the loaded area
d = mean effective slab depth = (dx + dy)/2
and where the design punching shear stress for nonprestressed slabs value is
vRdc = 0.12 k (100 ȡlfck )
1/3
(6.32b)
with
k = size factor = 1+ ¥(200/d) 2.0 d in mm
ȡl = ¥(ȡlx . ȡly) 0.02
fck = characteristic cylinder strength of concrete
In the new definition, according to MC’90, the control perimeter is moved from a distance 1.5d from
the column, to a distance of 2d, see Fig. 6.37. There are two reasons to adopt the distance 2d. First it
makes the limiting shear stress much more uniform for different column sizes. Second now for
punching the same formulation can be used as for normal shear in members without shear
reinforcement, where also Eq. 6.32b is applied.
Figure6.37. Basic control perimeters around loaded areas
Often questions are raised with regard to the coefficient 0.12 in Eq. 6.32b. Therefore at first an
evaluation is carried out in order to verify this value. Altogether 112 test results have been considered,
taken from [212]. 78 of those results refer to tests on specimens with cylinder strength ranging from
15 to 60 MPa, whereas 32 refer to tests on high strength concrete specimens with concrete cylinder
strengths ranging from 60 to 120 MPa. This enables a good evaluation of the validity of the punching
shear formula (Eq. 6.32a and 6.32b) for higher concrete strengths.
The tests cover the interval of individual parameters given in Table 6.9.
Table 6.9. Range of parameters in tests used for evaluation
Guide to EC2 Section 6
Page 627 Table of contents
Fig. 6.38 shows a diagram in which the values vexp /vcalc are shown as a function of the concrete
strength. The tests of Base fall somewhat outside of the scope probably due to the very small
maximum particle diameter (Dmax = 4,2 mm). Those tests therefore have to be regarded with some
reservation.
Figure6.38. Punching strength of slabs without shear reinforcement:
comparison of test results v/s eq.6.32
Fig. 6.39 shows the frequencies of the relative punching shear capacities (ratio experimental to
calculated values) for the data considered.
Figure 6.39. Frequencies of relative punching shear carrying capacity
Assuming a normal distribution, a mean value of 0.191 is obtained with a standard deviation of
į = 0.0247, and a coefficient of variation v = 0.13. Strictly speaking this would mean a characteristic
lower bound value of 0.191  0.164 ā 0.0247 = 0.15. assuming a safety factor of 1.5 this would result in
a coefficient of 0.10 in stead of 0.12 in the equation of the design punching shear stress (Eq. 6.32b).
However, although the variation of the concrete strength (for which a material safety factor 1.5
applies) is the dominating factor with regard to the scatter of results in punching shear tests, the
punching shear capacity is not linearly proportional to the concrete compressive strength (the fck
value has an exponent 1/3 in Eq. 6.32b ), so that simply applying a material safety factor 1.5 as well
for the derivation of the punching shear capacity would be inappropriate.
Therefore a more sophisticated approach was necessary in order to unambiguously derive a design
equation with the required level of reliability. For such a case the classical "level 2 method", as
described in ECl Basis of Design is suitable. The way how to deal with this method has been
described and illustrated by Taerwe [13]. The same method was applied by Konig and Fischer for
investigating the reliability of existing formulations for the shear capacity of members without shear
reinforcement [14].
According to the level 2 method, a reliable design equation can be derived from test results with the
general formulation
BRd = ȝBR (1  BR ȕ įBR) (6.33)
where
BRd design value
ȝBR mean value of test results
BR sensitivity factor for BR, normally taken as 0.8 in the case of one dominating parameter
ȕ target safety index, taken 3.8
įBR coefficient of variation
with ȝBR = 0.191, ĮBR = 0.8 and įBR = 0.130 a value for the design coefficient in Eq. 6.32b of 0.116
Guide to EC2 Section 6
Page 628 Table of contents
was obtained. In this derivation, however, the mean concrete cylinder compressive strength has been
used, whereas in the code expression the 5%lower value fck is used. In the Model Code the relation
fck = fcm  8 (MPa) (6.34)
is given. This means a coefficient of variation for a concrete C25 of v = 0.15 and for a concrete C90 of
v = 0.05. In combination with BR = 0.13 for the punching tests this would mean an increase of the
coefficient 0.116 of about 6.8 % for C25 and 3.6 % for C90 (see [14, p. 91). This would then result in a
coefficient 0.124 for a concrete class C25 to 0.120 for a concrete class C90.
It can therefore be concluded that equation 6.32a,b is correct.
A disadvantage of Eq. 6.32.a,b is that they go to 0 if µ goes to 0. This could give unrealistically low
punching shear capacities for low reinforcement ratio's, that may for instance occur in prestressed
slabs. Furthermore designers like to have a simple lower bound formulation for a fist check.
Therefore a lower bound was added, solely depending on the concrete tensile strength, according to
the relation vu > Cā fctd, where fctd = fctk /¸c. Evaluating the same results as shown in Fig. 6.38 with the
equation vu = Cā fctk, it was found that for normal strength concrete (<C50/60) a 5% lower value
C =0,57 applied and for high strength concrete (>C50/60) a value equal to C= 0,42. It should however
be noted that the collection of tests does not contain slabs with crosssectional depths larger than 275
mm. In order to cope with larger slab depths used in practice the coefficient C 5% should therefore be
further reduced. The value 0,35, used as a lower limit for shear as well, seems to be quite reasonable.
Taking into account this lower li the design equation for nonprestressed slabs should therefore be:
VRd,c = (0,18/¸c) k (100ȡl fck)
1/3
> 0,35fctd (6.35)
6.4.2.2 Punching shear resistance of prestressed slabs without shear reinforcement.
For shear loaded members the influence of a normal compression force is taken into account a
separate contribution of 0,15ıcp to the ultimate shear stress vRd.c (prENV 199211:2001 6.2a), see
also the report for shear. It is logic that also with regard to punching the effect of prestressing will be
positive. It is however not expected that the same term as for shear can be used, because the
contribution of prestressing to the punching resistance depends also decisively on the definition of the
control perimeter.
In order to find the contribution prestressing a selection of test results has been made: Andersson
[40], Gerber & Bums [41], Stahlton [42], Pralong, Brändli, Thürlimann [43] and Kordina, Nölting [44]. A
comparison of the test results with the design equation:
vRd,c = (0,18/¸c) k (100ȡl fck)
1/3
 0,08ıcp > 0,35 fctk /¸c  0,08ıcp (6.36)
The mean value of vexp/ vRd,c is 1,58 and the standard variation is s = 0,20. According to Eq. 6.33, with
Į = 0,8 and ȕ = 3,8 the design value should be (1,58  0,8 . 3,8 . 0,20) vRd,c = 0,972 vRd,c. Actually this
means that Eq. 6.36, giving values which are only slightly (2,8%) too high, is acceptable as a design
equation.
Fig 6.40. Verification of Eq. 6 with test results
6.4.2.3 Punching shear resistance of labs with punching shear reinforcement
In ENV 199211, Eq. 6.34, it was assumed that the contribution of punching shear reinforcement to
the total shear capacity can be accounted for by
ȈAsw fyd sin o (6.37)
Adding this contribution to the punching shear capacity of a similar slab without shear capacity,
Guide to EC2 Section 6
Page 629 Table of contents
according to Eq, 6.31, would gives the total punching shear resistance.
A first important question is if the summation principle of a concrete component and a punching
reinforcement component is valid anyhow. In the descriptive model of Kinnunen and Nylander [15] for
slabs without punching reinforcement the tangential compressive strain at the bottom face of the slab
is the design criterion. A punching shear reinforcement would then only be able to limit the rotation of
the kinematic punching shear mechanism and as such reduce the compressive strain in the critical
area, so that the failure load is increased. A further argument against the taking into account in full of
the reinforcement according to Eq. 6.34 is that it is hard to find adequately anchored punching shear
reinforcement at both sides of a critical crack Therefore the punching shear reinforcement is not yet
yielding when the concrete contribution is at its maximum. One can also say that due to the vertical
movement of the punching cone, concrete component has already a reduced value at the moment
that yielding of the shear reinforcement has been reached.
In literature two types of proposals are distinguished in order to cope with this phenomenon. One
group of researchers, such as Moe [15], Pranz [6], Herzog [16] Petcu [17], Kordina,/Nölting [18]
propose efficiency factors for the contribution of the shear reinforcement ranging from 0.80 down to
even 0.25. Others, like Elstner/Hognestadt [19],and Regan [20], propose to use the summation
principle, however with a reduced concrete contribution (efficiency factors ranging from 0.6 to 0.8).
Fig. 6.41 shows a comparison of test results (punching failures) from Gomcs [22, 23], Yitzakhi [25]
and Regan [24] with the formulation
Vu = 0.75Vc + Vs (6.38)
where Vc is the concrete contribution (punching shear capacity of similar slab without shear
reinforcement) and Vs is the contribution of the yielding steel.
In the prENV 199211:2001, the formulation according to MC’90 has been chosen, but with effective
design strength of the shear reinforcement which depends on the slab depth, in order to account for
the anchorage efficiency. This means that
VRd,cs = 0,75 vRd,cā ud + ȈAsw fywd,eff sinĮ (6.39)
where
vRd,c according to Eq. 6.32b
Į inclination of the shear reinforcement
fywd,eff design strength of shear reinforcement, according to fywd,eff = 250 + 0,25d < fywd (Mpa).
ȈAsw shear reinforcement within the perimeter considered.
Figure 6.41. Punching capacities related to resistance of the shear reinforcement (Regan [24])
The increase of the punching shear capacity by shear reinforcement is limited to a bound. In the
version of EC2 of 1988 the upper bound was formulated as vRd,max = 1.4 vRd1, where vRd1 follows from
Eq. 6.31b. In a redraft in 1991 this value was increased to 1.6 vRd1. This upper limit will, however,
hardly be reached in practical situations, where normal punching shear reinforcement is used.
However, using shear heads and shear studs, which can be more efficient than stirrups as shear
reinforcement, higher values of the upper punching capacity than 1.6vRd1 can be reached. Therefore a
Guide to EC2 Section 6
Page 630 Table of contents
modified upper limit has been defined according to MC’90, according to
VRd,max = 0.5v fcd u0 d (6.40)
where u0 is the length of the column periphery, and v is equal to
v = 0.60 (1fck /250) (6.41)
The distance from the column to the inner shear reinforcement should not be larger than 0.5d, nor
should it be less than 0.3d, since steel closer than this will not be well anchored in the compression
zone if intersected by cracks at lower inclinations (Regan [24]). The distance between the layers of
shear reinforcement in radial direction should not be larger than 0.75d.
It should be verified that no punching failure occurs outside the outermost layer of shear
reinforcement. Therefore an additional perimeter un is defined at a distance of 1,5d from the
outermost shear reinforcement. It should be shown that here
vEd < 0.12 k (100ȡl fck)
1/3
 0.08ıcp (6.42)
where vEd is the design calculated of the ultimate shear stress on the perimeter un, see fig. 6.42
Figure 6.42. Control perimeter un at interior column
Practically the design of the shear reinforcement is quite simple. At first at the perimeter with distance
2d from the loaded area the punching capacity is checked and the eventual shear reinforcement is
calculated. Then the perimeter is determined for which vEd < vRd,c.
Finally the shear reinforcement (with the same cross section per unit area) is extended to a distance
1,5d from the outer perimeter.
6.4.3. Basic equation for eccentric punching
In the draft of 1988 only a very general approach as in combination with a bending moment. The
design punching shear stress vEd was formulated as
vEd = (ȕVEd)/(u.d) (6.43)
where
VEd design value of the punching shear
ȕ factor taking account of the expected effect of eccentricity, “in the absence of a more
rigorous analysis”. No further indication on what this “more rigorous analysis” means was
given. In Fig. 6.43 the values for the eccentricity factor ȕ are given.
Figure 6.43. Approximate values for eccentricity factor  in new draft
Complementary to this simplified approach the more accurate method, given in MC’90 has been
adopted in prENV 199211:2001. This method takes the effect of an unbalanced moment into
account with the formulation:
VEd = VEd/(u1d) + (KMEd)/(W1d) (6.44)
where W1 is a function of the control perimeter u1:
1
u
1 0
W e d =
)
A
The property W1 corresponds to a type of “plastic” distribution of the shear stresses as illustrated in
Guide to EC2 Section 6
Page 631 Table of contents
Fig. 6.44. An analysis by Mast [34,35] on the basis of an elastic analysis of the distribution of shear
stresses in a slab in the vicinity of a column showed that those stresses approach the distribution
shown in Fig. 6.45 quite well. For a rectangular column W1 follows from
W1= c1
2
+ c1 c12 + 4c2d + 17.8d
2
+ 2tdc1 (6.45)
Figure 6.44. Shear distribution due to an unbalanced moment at a slab internal column connection
K is a factor taking into consideration that a bending moment in the slab, is only sustained by bending
in the column but also by bending and torsion in the slab itself. K follows from the table below
For round columns c1/c2 = 1 so K = 0.6.
With this approach the shear capacity depends on the column size and the value of the unbalanced
moment and is therefore much more accurate than from Eq. 6.40.
Fig. 6.45 shows Eq. 6.41 in comparison with tests results. The diagram is a combination of two
figures, from Regan [24,36].
Fig 6.45. Comparison of Eq. 11 with test on interior column – slab connections
The method can be as well applied to edge and corner columns. However, in those cases the ultimate
punching shear stress can as well be calculated in a simplified way assuming uniform shear on the
reduced perimeter shown in Fig. 6.46.
Figure 6.46. Reduced perimeters for assumed uniform shear for edge and corner columns
Guide to EC2 Section 6
Page 632 Table of contents
6.5 Design with strut and tie
models
6.6 Anchorages and laps
6.7 Partially loaded areas
6.8 Fatigue
6.4.3.1 Punching shear in column bases
The most important difference between punching of a slab around a column and punching of column
base supporting a column is the presence of a significant counter pressure from the soil. A second
difference is that the distance from columns to the edges of their bases are commonly much smaller
than those to sections of radial contra flexure in suspended slabs [38].
Due to the influence of the vertical soil pressure, the inclination of the punching cone in column bases
may well be steeper than in suspended slabs, which gives rise to uncertainty with regard to the critical
perimeter to be checked. This was not regarded in the EC2 version of 1988. The CE Model Code
1990 gives an alternative, in which the position of the control perimeter is treated as a variable and
the unit punching resistance, is taken to vary with the distance from the column to the control
perimeter, i.e. with the inclination of the failure surface.
For concentric loading the design punching force is
VEd,red = VEd  ǻVEd (6.46)
where
VEd column load
ǻVEd the upward force within the control perimeter considered i,e. upward pressure from soil
minus self weight of base
vEd = VEd,red/ud (6.47)
where u is the control perimeter taking a value a < 2d instead of 2d into account (see also fig. 6.37).
The nominal ultimate shear stress at the perimeter is
vRdc = 0,12 k(100ȡfck )
1/3
2d/a < 0.5vfcd (6.48)
where
a distance from the periphery of the control perimeter considered
v 0.60 (1  fck /250) (6.49)
Fig. 6.47 shows a result of a parameter study, carried out with the previous equations. For many
combinations of base width to column width l/c and column width to effective slab depth c/d the critical
ratio acrit /d has been determined, for which the lowest column load is obtained. The results are shown
in Fig. 6.47. It turns out that the ultimate column load is a function of the based to column width l/c but
is independent of the ratio c/d. In the lower diagram of fig. 6.47 the design column load VEd can
immediately be determined as a function of the ratio l/c. The corresponding value of the critical control
perimeter acrit /d is read in the upper diagram. It can be seen that in the utmost number of cases the
value of acrit is smaller than 2d, which means indeed that the inclination of the punching cone is much
steeper than in suspended slabs.
Figure 6.47. Shear capacity of column bases
See example n. 6.8 to 6.14
See example n. 6.15
No comments
No comments
Guide to EC2 Section 6
Page 633 Table of contents
REFERENCES
Shear
Aparicio,A., Calavera, J., del Pozo, F.J.,(2000) “Testing strut compression shear failure in beams”,
Polytechnic University of Barcelona.
Asin, M. (2000), “The behaviour of reinforced concrete deep beams”, PhDThesis, Delft University of
Technology, The Netherlands.
Aster, H., Koch, R. (1974), “Schubtragfähigkeit dicker Stahlbetonplatten”, Beton und Stahlbetonbau,
11/1974
Baldwin, J.W ., Viest, I.M. (1958), “Effect of axial compression in shear strength of reinforced concrete
frame members”, ACIJournal, Vol. 30, No. 4, Nov. 1958, pp. 635654.
Bennett, E.W., Balasoorya, B.M.A.(1971),”Shear strength of prestressed beams with thin webs failing
in inclined compression”, ACIJournal, March 1971.
Bennett,E.W. ,Debaiky, S.Y. (1974),”High strength steel as shear reinforcement in prestressed
concrete beams”, ACI Special Publication 42, Vol. I, pp. 231248, 1974.
Bernaert, S., and Siess, O.(1956), “Strength of shear in reinforced concrete beams under uniform
loading”, University of Illinois, June 1956.
Bhide, S.B., Collins, M.P.,(1989)’’Influence of Axial Tension on the Shear Capacity of Reinforced
Concrete Members”, ACI Journal Sept.Oct. 1989, pp. 570581.
Chen Ganwei (1988), “Plastic Analysis of Shear in Beams, Deep, Beams and Corbels, Department of
Structural Engineering, Technical; University of Denmark, Serie R, No. 237, 1988.
Clark, A.P. (1951), “Diagonal tension in reinforced concrete beams”, ACI Journal Vol. 32, No. 2,
October 1951, pp. 145156.
De Cossio, R.D., Siess, C.P. (1960), “Behavior and strength in shear of beams and frames without
web reinforcement”, ACIJournal, Feb. 1960, pp. 695735.
Fouré, B. (2000), “Proposal for rewriting item 6.2.4 in section 6.2 (Ultimate Limit State – Shear”, Note
to Project Team for EC2, May 2000 (unpublished).
Görtz, S., Hegger, J (1999), “Shear Capacity of Prestressed Concrete Beams of HSC Elements”,
Symposium Utilization of High Strength/ High performance Concrete, Sandefjord, Norway, 2024 June
1999, proceedings, Vol. 1, pp. 312321.
Hamadi, Y.D. and Regan, P.E. (1980), “Behavior of normal and lightweight aggregate beams with
shear cracks”, The Structural Engineer, Vol. 58B No. 4, December 1980, pp. 7179.
Hanson, J.M., Hulsbos, C.L. (1964),”Ultimate shear tests on prestressed concrete Ibeams under
concentrated and uniform loading”, PCIJournal, June 1964, pp.1528
Hedman, O., Losberg, A. (1978), “Design of concrete structures with regard to shear forces”, CEB
Bulletin d‘Information No. 126, pp. 184209, 1978.
Hognestadt, E. ,Elstner, R.C. (1957), “Laboratory investigation of rigid frame failure”, ACI Journal, V.
53 No. 1, Jan 1957, pp. 637668.
Kani, G.N.J. (1966), ”Basic Facts Concerning Shear Failure”, ACIJournal 64, (1966), No. 6,
pp. 675 – 692.
König, G., Fischer, J. (1995), “Model Uncertainties concerning Design Equations for the Shear
Capacity of Concrete Members without Shear Reinforcement”, CEB Bulletin 224, July 1995,
pp. 49100.
Krefeld, W.J., and Thurston, C.W. (1966), “Studies of the shear and diagonal tension strength of
simply supported reinforced concrete beams”, ACI Journal, Vol. 63, No. 4, April 1966, pp. 451476.
Küng, R., “Ein Beitrag zur Schubsicherung im Stahlbetonbau”, Betonstahl in Entwicklung, Heft 33,
TorEsteg Steel Corporation, Luxemborg.
Lehwalter, N. (1988), “The bearing capacity of concrete struts in strut and tie models exemplified for
the case of short beams”, PhDThesis, Darmstadt University of Technology , Germany (in German).
Leonhardt, F., Walther, R. (1961), “The Stuttgart shear tests 1961”, Series of papers in Beton  und
Stahlbetonbau,
Leonhardt, F., Koch, R., Rostsy, F., (1973), “Schubversuche an Spannbetonträgern”, Deutscher
Ausschuss für Stahlbeton, Heft 227, Berlin.
Levi, F., Marro, P. (1993), “Shear tests on HSC prestressed beams – proposals of new interpretative
models”, Conference on High Strength Concrete 1993, Lillehammer 20 24 June, Proceedings, pp.
293305.
Guide to EC2 Section 6
Page 634 Table of contents
Leung, Y.W., Chew, C.B. and Regan, P.E. (1976), “Shear strength of various shapes of concrete
beams without shear reinforcement”, Polytechnic of Central London, 1976.
Lyngberg, B.S., (1976), Ultimate shear resistance of partially prestressed concrete Ibeams”,
ACIJournl, April 1976, pp. 214221.
Mathey, R.G., Watstein, D. (1963), “Shear strength of beams without web reinforcement containing
deformed bars of different yield strengths”, ACI Journal Vol. 60., No. 2, Feb. 1963, pp. 183207.
Moayer, H. and Regan, P.E., (1974) ”Shear strength of prestressed and reinforced concrete
Tbeams”, ACI Special Publication 42, Vol 1, pp. 183221, 1974.
Morrow, J., Viest, .M. (1957), “Shear strength of reinforced concrete beams, ACI Journal, Vol. 28, No.
9, March 1957, pp. 833869.
Muhidin, N.A. and Regan, P.E. (1977), “Chopped steel fibres as shear reinforcement in concrete
beams”, Fibre reinforced materials – design and engineering applications, Proceedings of a
conference, Institution of Civil Engineers, London 1977, pp. 135  149.
Nielsen, M.P. (1990), “Commentaries on Shear and Torsion”, Eurocode 2 editorial Group – 1st draft –
October 1990.
PrEN 1990, Eurocode: Basis of Structural Design, Stage 34, Draft version Dec. 2001.
Regan, P.E. (1971), “Shear in reinforced concrete – an experimental study”, Technical Note 45,
CIRIA, London.
Regan, P.E. (1999), “Ultimate Limit State Principles: Basic design for moment, shear and torsion”, in
fib Text Book on Structural Concrete, Vol. 2, pp. 141223.
Regan, P.E., RezaJorabi, H. (1987), “The shear resistance of reinforced concrete Ibeams”, Studi e
Ricerche, Politecnico di Milano, Vol 19, 1987, pp. 305321.
Regan, P.E. (1998),”Enhancement of shear resistance in short shear spans of reinforced concrete”,
an evaluation of UK recommendations and particularly of BD 44/95, University of Westminster,
London, 1998 (14 pages).
Regan, P.E. (2000), “Aspects of diagonal tension in reinforced concrete”, Structural Concrete, Journal
of fib, No. 13, 2000, Sept. 119132.
Rogowski, D.M., MacGregor, J.G. (1983),”Shear strength of deep reinforced concrete continuous
beams”, Structural Engineering Report, No. 11., University of Alberta, Nov. 1983.
Rüsch, H., Haugli, F.R., Mayer, H. (1962), “Shear tests on rectangular reinforced concrete beams with
uniformly distributed loads”, Deutscher Ausschuss für Stahlbeton, Berlin, Heft 145 (in German).
Sörensen, H.C. (1974). “Shear tests on 12 reinforced concrete Tbeams”, report R60, Structural
research Laboratory, Technical University of Denmark, 1974.
Taerwe, L.R. (1993), “Toward a consistent treatment of model uncertainties in reliability formats for
concrete structures”, CEB Bulletin d’Ínformation 219, “Safety and Performance Concepts”, pp. 535.
Walraven, J.C. (1978), “The influence of member depth on the shear strength of normal and
lightweight concrete beams without shear reinforcement”, Stevin Report 5784, Delft University of
Technology, 1978.
Walraven, J.C. (1987), “Shear in prestressed concrete”, CEBBulletin d’Information Nr. 180, 144
pages.
Walraven, J.C., Lehwalther, N. (1989), “The bearing capacity of concrete struts in strut and tie models
exemplified for the case of short beams”, Betonund Stahlbetonbau, Heft 4, April ’89,
pp. 8187 (in German).
Walraven, J.C., Lehwalther, N. (1994), “Size effects in short beams loaded in shear”, AC_  Structural
Journal, Vol. 91., No. 5., Sept.Oct. 1994, pp. 585593.
Walraven, J.C., AlZubi, N. (1995),”Shear capacity of lightweight concrete beams with shear
reinfocement”, Proceedings Symposium on Lightweight Aggregate Concrete, Sandefjord, Norway,
Vol. 1, pp. 91104
Walraven, J.C., Stroband, J. (1999), Shear capacity of high strength concrete beams with shear
reinforcement”, Symposium on High Strength Concrete, Sandefjord, Norway, 2024 June 1999,
Proceedings Vol. 1. pp. 693700.
Guide to EC2 Section 6
Page 635 Table of contents
Punching
1. Walraven, J.C.; CEBBulletin d'lnformation 180, “Shear in Prestressed Concrete” pp.6871.
2. Kinnunen, S., H. Nylander; “Punching of concrete slabs without shear reinforcement”,
Transactions of the Royal Institute of Technology, Stockholm, Civil Engineering 3,Nr. 158, 1960.
3. Marti, P, J. Pralong, B. Thurlimann; “Schubversuche an Stahlbetonplatten”, Institut fur Baustik
und Konstruktion, ETH Zurich, Berich Nr. 73052, Birkhauser Verlag Basel, September 1977.
4. Seible, F., A. Ghali, W.H. Digler; “Preassembled shear reinforcing units for flat plates”, ACI
Journal, JanFeb. 1980, pp. 2835
5. Pralong, J., w. Briindli, B. Thurlimann; "Durchstanzversuche an Stahlbetonund
Spannbetonplatten", Institut fur Baustatik und Kkonstruktion, EHT ZUrich, Bericht Nr. 73053,
Birkhiiuser Verlag Basel, December 1979.
6. Franz, G.; "Versuche an Stahlbetonkorper der Flachdecke im StiitzenbereichVersuchsreihen I en
II, Institut fur Beton und Stahlbetonbau," TH Karlruhe, 1963 und 1964.
7. Base, G.D.;CEBBulletinNr.57, 1968,pp.6882
8. Tolf, P.; "Effect of slab thickness on punching shear strength of concrete slabs. Tests on circular
slabs", Bulletin 146, Dept. of Structural Mechanics and Engineering, Royal Institute of
Technology, Stockholm (in Swedish with summary in English).
9. Marzouk, H., A. Hussein; "Experimental investigation on the behavior of high strength concrete
slabs" ACIStructural Journal, Vol. 88, No.6, American Concrete Institute, Detroit, pp. 701713.
10. Tomaszewicz, A.; "HighStrength .SP2 Plates and Shells, Report 2.3 "Punching shear capacity
of reinforced concrete slabs", Report No. STF70 A93082, SINTEF Structures and Concrete,
Trondheim, 36 pp.
11. Regan P.F., A. AlHussaini, K.E. Ramdane, H.Y. Xue; "Behavior of high strength concrete slabs",
Proceedings of the international conference Concrete 2000, Sept. 79, 1993, Vol. 1, pp. 761773.
12. Hallgren, M; "Punching shear tests on reinforced high strength concrete slabs", KTH Stockholm,
Technical Report 1994, No.14. Structural Mechanics.
13. Taerwe, L.R.; "Towards a consistent treatment of model uncertainties in reliability formats for
concrete structures", CEBBulletin d'Information 219, "Safety and Performance Concepts",
p 535.
14. Konig, G., Fischer, J,,"Mode1 Uncertainties concerning Design Equations for the Shear Capacity
of Concrete Members without Shear Reinforcement", in CEB Bulletin 224, Model Uncertainties,
July 1995, pp. 49100.
15. Kinnunen, S., Nylander, H., "Punching of Concrete Slabs without Shear Reinforcement"
Meddelande N. 38, Institutionen for Byggnadstatik, Kungilga Tekniska Hogskolan, Stockholm
1960.
16. Moe, J .,"Shearing strength of reinforced concrete slabs and footings under concentrated loads"
Development Department Bulletin d47, Portland Cement Association, Apri11961.
17. Herzog, M.,"Der Durchstandswiderstand von Stahlbetonplatten nach neu ausgewerteten
Versuchen", Ostereichische Ingenieurzeitschrift, Vol. 14, 1971, Nr. 6, pp. 186192, Nr. 7.
Pp.216219.
18. Petcu, V., Stanculescu, G., Pancaldi, U., "Punching strength predictions for twoway reinforced
concrete slabs", Revue Romaine des Sciences Techniques: serie de la mechanique applique, No
2, March/Apri11979, Rumania.
19. Kordina, K., No1ting, D., "Shear capacity of reinforced concrete s1abs subjected to punching",
Deutscher Ausschuss fur Stahlbeton, Nr. 371, Berlin, 1986 (in German)
20. Elstner, R.C., Hognestad, E., "Shearing strength of reinforced concrete slabs", ACI Journal, Vo1.
53, July 1956, pp. 2958.
21. Regan, P.E., "Single legged stirrups as shear reinforcement in reinforced concrete flat slabs",
Structures Research Group, Polytecnic of Central London, 1980
22. Gomes, R.B.,"Punching resistance of reinforced concrete flat slabs",PhDThesis, Polytechnic of
Central London, 1991
23. Gomes, R.B., Andrade, M.A.S.,"Punching in reinforced concrete flat s1abs with holes",
Developments in Computor Aided Design and Modeling for Structural Engineering, Civil Comp
Press, , Edingburgh, pp. 185193.
24. Regan, P.E., "Ultimate Limit State Principles", Model Code Text Book, Part 7.4 Punching, in fib
Guide to EC2 Section 6
Page 636 Table of contents
Bulletin 2, Structural Concrete, July, 1999, Vol. 2, pp. 202223.
25. Yitzhaki, R., "Punching strength of reinforced concrete slabs", ACI Journal Vol. 63, Nr. 5, Mau
1966, pp. 527542.
26. Stamencovic, A., Chapman, J.C., "Local strength of flat slabs at column heads",
27. Report Nr. 39, Construction Industry Research and Information Association, London, 1972.
28. Shehata, L.A.E., "Puncionamento assimetrico em lajes de concreto, COPPE/UFRJ, Rio de
Janeiro, 1987
29. Regan, P .E., "Behaviour of reinforced concrete flat slabs", Report No.89, Construction Industry
Research and Information Association, London 1980.
30. Hanson, N.W., Hanson, J.M., "Shear and moment transfer between concrete slabs and
columns", Portland Cement Association, Development Bulletin D129, Jan 196B.
31. Islam, S., Park, R.,"Tests on slabcolumn connections with shear and unbalanced flexure,
Journal of the Structural Division, Proceedings, ASCE, Vol. 1002 No ST3, March 1976,
pp. 549568.
32. Anis, N.N.,"Shear strength of reinforced concrete flat PhD slabs without shear reinforcement",
Thesis Imperial College, London, 1970.
33. Narasimhan, N., "Shear reinforcement in reinforced concrete column heads", PhDThesis,
Imperial College, London 1971
34. Hawkins, N.M., Bao, A., Yamazaki,J.,"Moment transfer from concrete slabs to columns", ACI
Structural Journal, v. 86, No.6, NovDec. 1998, pp. 705716.
35. Mast, P.E., "Stresses in flat plates near columns", ACI Journal, Oct. 1970, pp. 761 768.
36. Mast, P.E.,"Plate stresses at columns near to free edges", ACI Journal, Nov. 1970, pp. 899 902.
37. Regan, P .E., "Explanatory notes on proposed Model Code (MC'90)", Note to CEB Commission
IV, May 8th 1990.
38. Regan. P.E., Braestrup. M.W., "Punching shear in reinforced concrete", CEBBulletin
d'Information No.168, Jan.1985.
39. Lucio, V.J.G., Appleton, J.A.S., Almeida, J.P., "Ultimate limit state of punching in the (fib) FIP
Recommendations for the design of posttensioned slabs and foundations", Structural Concrete,
Journal of fib, 2000, 1, No.3, Sept. 143149.
40. Andersson, J.L., "Genomstansning ay lift slabs (Punching of lift slabs)", Nordisk Betong,
Year 7,1963, H.3, pp. 229252.
41. Gerber, L.L., Burns, N.H., "Ultimate strength tests of posttensioned flat plates", PCIJournal,
Nov/Dec. 1971, pp. 4058.
42. Stahlton, A.G., "Flat slabs with Stahlton column strip prestressing", unpublished Report of
Stahlton AG, Zurich, 1974/75 (in German).
43. Pralong, J., Brandli, W., Thurimann, B., Punching tests on reinforced and prestressed slabs",
Institute for Building Statics and Structural Design", ETH Zurich, Report Nr.73053, Birkhiiuser
Verlag Basel, December 1979 (in German).
44. Kordina, K., Nolting, D.,"Tests on shear in slabs, prestressed with unbonded tendons", Report of
the Institute for Materials, Concrete Structures and Fire, TU Braunschweig, September 1984
(in German).
Guide to EC2 Section 7
Page 71 Table of contents
SECTION 7. SERVICEABILITY
LIMIT STATES (SLS)
7.1 General
7.2 Stress limitation
7.3 Crack control
7.3.1 General considerations
7.3.2 Minimum reinforcement
areas
SECTION 7. SERVICEABILITY LIMIT STATES
See example n. 7.1
See example n. 7.3
C7.3.2 Formula for minimum reinforcement
Formula 7.1 of prEN can be partially deduced by expressing the condition that the minimum
reinforcement should be able to withstand the cracking moment working at a certain stress ıs fyk.
Figure 7.1 Minimum reinforcement concept
For a rectangular cross section subject to combined bending and axial force, the cracking moment
can be determined by:
,
1
3 3 2
Ed
cr ct eff ct cr
N h
M f A h h
 
= + +

\ .
(7.1)
After cracking this same bending moment, together with the axial force, must be taken by the cracked
cross section:
( )
,
1
0,9
3 3 2 2
Ed
ct eff ct cr s s Ed
N h h
f A h h A z N h z
   
÷ + = o ÷ ÷ ÷
 
\ . \ .
(7.2)
The lever arm z, may be taken as a certain fraction o of h (around 0.8h for pure bending). Introducing
the following notation:
ct cr
A bh =
o =
Ed
c
N z
bh
= oh
Equation above can be rewritten as:
( )
,
,
1
1 1 3 0,9 1
3
c
ct eff ct s s
ct eff cr
h
f A A
f h
  ( o
( ÷ + ÷o ÷ = o

(
¸ ¸

o
¸ ¸ \ .
(7.3)
, ct eff ct c s s
f A k A = o (7.4)
The above value of kc is valid only if o > 0, since the deduction assumes that there is a part of the
cross section which is in compression. In order to approximate kc in the following figure the value of
o=z/h is plotted against the adimensional axial force
Ed
ck
N
bhf
v = within a range of +0.03 to –0.22. Over
0.03, there is no compression block and under –0.22 there is no need to provide minimum
reinforcement according to the principle established above.
Factor alpha=z/h as a function of adimensional axial force
Figure 7.2 Value of o as a function of the axial force
It can be seen from the above figure that it is reasonable to assume a value of 0.8 for parameter o.
With this assumption and for compression force or moderate tensile force, the values of kc may be
estimated as:
,
0,4 1 1 0,7
c
c
ct eff cr
h
k
f h
    o
= ÷ ÷
 

\ . \ .
(7.5)
hcr can be computed as a function of the axial compression and the tensile strength of concrete.
Guide to EC2 Section 7
Page 72 Table of contents
The cracking moment is given by the following condition:
( )
2
, , 2 2
6 6
6
Ed cr cr
ct eff c cr ct eff c
N M M bh
f M f
bh bh bh
= + × = o + × ¬ = ÷o (7.6)
The depth of the tensile zone prior to cracking, relative to the centre of gravity of the cross section, is
given by the condition that the stress be nil when the existing axial force is applied together with the
cracking moment:
( )
3
,
0 12
2
cr c
c
ct eff c
M h
x x
bh f
o
= o + ¬ = ÷
÷o
(7.7)
Therefore, hcr can be calculated as:
( )
( )
,
, ,
1 2
2 2
ct eff c
c
cr
cr ct eff ct eff c
f
h h h
h x
h f f
    ÷o
o
= ÷ = + ¬ =  
 
÷o
\ . \ .
(7.8)
Introducing equation (7.8) in (7.5), the expression of kc becomes:
,
, ,
0,4 1 1 1,4
ct eff c c
c
ct eff ct eff
f
k
f f
(     ÷o o
= ÷ ÷ (  
 
(
\ . \ . ¸ ¸
(7.9)
In the following figure the value of kc is plotted against ıc for a value of fct,eff of 2.9 Mpa according to
both the above formulation and the equation of prEN.
Figure 7.3. Value of kc as a function of the axial force
In the next figure the minimum reinforcement for a section subject to a compressive force (expressed
in terms of a reduced axial force) is shown calculated according to 4 different methods:
 Proposal of prEN
 Compensation of tension block. This procedure is on the safe side since it neglects the
increase of the lever arm which occurs after cracking.
 Direct calculation (exact determination of o)
 Procedure explained above in which Į is taken as a constant equal to 0.8.
Figure 7.4. Comparison of different approximate expressions for the calculation of the
minimum reinforcement
It can be seen that the 2 last methods give almost the same results. It can also be seen that the
formula of prEN is mostly on the safe side.
Guide to EC2 Section 7
Page 73 Table of contents
7.3.3 Control of cracking
without direct calculation
C7.3.3. Formula for maximum bar diameter
In this paragraph the formula for the maximum bar diameter is deduced for three crack width
formulations: MC90 [3], EC2 [4] and prEN. This deduction, as will be seen, requires the introduction
of simplifications and the assumption that the steel ratio, for which a certain stress ıs is achieved
corresponds to the minimum steel ratio given by prEN equation 7.1. This assumption is on the safe
side. A more exact formulation could be obtained if ȡeff were to remain a variable.
C7.3.3.1 Deduction of formula
Using the MC90 formulation, the crack width is given by:
( ) 2 1
3,6
s sr
k t sm cm t
eff s s
w l k
E
  o o 
= c ÷c = ÷

µ o
\ .
(7.10)
ısr is the stress calculated for the fully cracked cross section for the cracking moment. ısr may be
calculated by:
( )
, ,
,
2,5 '
ct eff ct c ct eff c cr
sr
s s eff
f A k k f k k h
A k h d
o = =
µ ÷
(7.11)
In the above expression, k’ is a factor which is equal to 1 for bending and equal to 2 for tension.
Introducing the expression of ısr into equation (7.10), and rearranging:
( )

 

\ .
k eff
ct,eff c cr
t
s,eff s
3, 6 w ȡ
=
f k h
1 k
ȡ ı 2, 5k' h d
(7.12)
In order to obtain numerical values from this expression, it is necessary to assume values for those
coefficients which are not considered as variables in table 7.1. The following values have been
assumed to derive table 7.1:
k = 1,0 ĸ h 0,3 (assumption on safe side)
kc = 0,4; k’ = 1 (pure bending)
kt = 0,38
~
cr
h h 10
= 5
h d 2 h
(7.13)
fct,eff = 2,9 N/mm
2
wk = 0,3 mm
With the above values equation (7.12) can be written as:

 

\ .
k eff
s
s,eff s
720000 w ȡ 1
=
ı
1
1 0,88
ȡ ı
(7.14)
The above expression, which is valid only for ıs > ısr, is a function of 3 parameters. Therefore some
assumption regarding ȡeff must be made in order to obtain table 7.2. The assumption which will be
made is that ȡeff =ȡeff,min (i.e. ıs = ısr). This assumption is justified because lower values of ıs will not
produce cracking and because if cracking does occur, then formula (7.14) will give smaller bar
diameters if the smallest the value of ȡeff is used. The value of ȡeff can therefore be taken from the
following equation:
( )
,
,min
2,5 '
ct eff c cr
eff eff
s
f k k h
k h d
µ = µ =
o ÷
(7.15)
If the above value of ȡeff is substituted into Eq. (7.12), this equation can be simplified into:
( ) ( )
,
2 2
3,6 808258
1 2,5 '
k ct eff c s cr
t s s
w f k k E h
k k h d
 = ~
÷ o ÷ o
(7.16)
In Table 1 the minimum reinforcement ratio which can to be used in equation (7.14) is given for each
stress level. In the same table, the maximum bar diameter obtained using equation (7.14) for wk=0.3
mm and that included in MC90 Table 7.4.3 are also given.
Table 7.1. MC90  Minimum reinforcement ratio for a given value of ıs which fulfils the condition
ıs > ısr. Value of corresponding max
Guide to EC2 Section 7
Page 74 Table of contents
These results are plotted in the following figure.
Figure 7.5. Theoretical and actual values of Dmax, according to MC90
Using the EC2 formulation, the crack width is given by:
( )
 
    
 c c
 

\ . \ .
\ .
2
s sr
k rm sm cm 1 2 1 2
eff s s
ı ı
w = 1,7s  = 1, 7 50+ 0, 25k k 1 ȕ ȕ
ȡ E ı
(7.17)
ısr is the stress calculated for the fully cracked cross section for the cracking moment. ısr may be
calculated by:
( )
ct,eff ct c ct,eff c cr
sr
s s,eff
f A k k f k k h
ı = =
A ȡ 2,5k' h d
(7.18)
Introducing the expression of ısr into equation (7.10), and rearranging:
( )
¦ ¹
¦ ¦
¦ ¦
¦ ¦

´ `
 
  ¦ ¦


¦ ¦

\ .
¦ ¦
\ . ¹ )
s eff k
2
s 1 2
ct,eff c cr
1 2
s,eff s
E ȡ w
=  50
1, 7ı 0,25k k
f k h
1 ȕ ȕ
ȡ ı 2,5k' h d
(7.19)
In order to obtain numerical values from this expression, it is necessary to assume values for those
coefficients which are not considered as variables in table 7.2. The following values have been
assumed to derive table 7.2:
k = 1,0 ĸ h 0,3 (assumption on safe side)
kc = 0,4; k’ = 1 (pure bending)
~
cr
h h 10
= 5
h d 2 h
(7.20)
fct,eff = 2,5 N/mm
2
12 = 0,50
0,25 k1k2 = 0,10
With the above values equation (7.19) can be written as:
¦ ¹
¦ ¦
¦ ¦
¦ ¦

´ `
 
  ¦ ¦


¦ ¦

\ .
¦ ¦
\ . ¹ )
k
eff
2
s
s,eff s
w 117647
=  50 10ȡ
ı
1.414
1
ȡ ı
(7.21)
Similarly, a value must be given to ȡeff in order to obtain values from this expression. As stated before,
it is on the safe side to assume for this purpose that ıs = ısr and that therefore the value of ȡeff can be
determined using equation (7.15). With this assumption, equation (7.19) can be rewritten as:
( ) ( )
¦ ¹ ¦ ¹
¦ ¦
 ~
´ ` ´ `
¦ ¦ ¹ ) ¹ )
ct,eff c s cr k
1 2 s 1 2 s s
f k k E h w 70588 20
=  50  50
1 ȕ ȕ 1, 7ı 0,25k k 2, 5k' h d ı ı
(7.22)
In the following table, the results obtained by application of the above formula and the values of ĳmax
according to EC2 are compared.
Guide to EC2 Section 7
Page 75 Table of contents
Table 7.2 EC2  Minimum reinforcement ratio for a given value of ıs which fulfils the condition
ıs > ısr. Value of corresponding max
In the following graph, the above results are plotted.
Fig 7.6. Theoretical and actual values of Dmax, according to EC2
Using the prEN formulation, the crack width is given by:
( )
      
c c
  

\ . \ . \ .
s sr
k rm sm cm 1 2 t
eff s s
ı ı
w = 1,7s  = 3, 4c + 0, 425k k 1 k
ȡ E ı
(7.23)
ısr is the stress calculated for the fully cracked cross section for the cracking moment. ısr may be
calculated by:
( )
ct,eff ct c ct,eff c cr
sr
s s,eff
f A k k f k k h
ı = =
A ȡ 2,5k' h d
(7.24)
Introducing the expression of ısr into equation (7.23), and rearranging:
( )
¦ ¹
¦ ¦
¦ ¦
¦ ¦

´ `
 
  ¦ ¦


¦ ¦

\ .
¦ ¦
\ . ¹ )
s eff k
2
s 1 2
ct,eff c cr
t
s,eff s
E ȡ w
=  3, 4c
ı 0, 425k k
f k h
1 k
ȡ ı 2, 5k' h d
(7.25)
In order to obtain numerical values from this expression, it is necessary to assume values for those
coefficients which are not considered as variables in table 7.2. The following values have been
assumed to derive table 7.2:
k = 1,0 h 0,3 (assumption on safe side)
kc = 0,4; k’ = 1 (pure bending)
~
cr
h h 10
= 5
h d 2 h
(7.26)
fct,eff = 2,9 N/mm
2
kt = 0,40
0,425 k1k2 = 0,17
C = 25 mm
With the above values equation (7.25) can be written as:
Guide to EC2 Section 7
Page 76 Table of contents
¦ ¹
¦ ¦
¦ ¦

´ `
   
¦ ¦
 
¦ ¦

\ . \ . ¹ )
k
eff
s
s,eff s
w 200000
=  85 5,88ȡ
ı
0, 928
1
ȡ ı
(7.27)
The above expression is a function of 3 parameters and again can be simplified with the assumptions
explained above, into:
( ) ( )
¦ ¹ ¦ ¹
¦ ¦
 ~
´ ` ´ `
¦ ¦ ¹ ) ¹ )
ct,eff c s cr k
t s s 1 2 s s
f k k E h w 100000 13, 65
=  3, 4c  85
1 k ı ı 0, 425k k 2, 5k' h d ı ı
(7.28)
This curve is represented in the following figure and given in numerical form in table 7.3. It can be
seen that good agreement is obtained between theory and prEN table. The small differences
observed are due to the need to use commercial bar diameters in table 7.3.
Table 7.3 prENV  Minimum reinforcement ratio for a given value of ıs which fulfils the condition
ıs > ısr. Value of corresponding max
Figure 7.7. Theoretical and actual values of Dmax, according to prEN
The three formulas analyzed are compared in the following graph. It can be seen that the theoretical
results for EC2 and prEN are very similar while the MC90 equation is somewhat more conservative.
The values included in the tables of the codes do not always match the theoretical equation but are
generally conservative for EC2 and prEN. The MC90 table values are somewhat above the theoretical
curve as already shown before.
Figure 7.8 Theoretical and actual values of Dmax. Comparison of the 3 models considered
Guide to EC2 Section 7
Page 77 Table of contents
7.3.4 Calculation of crack
widths
7.3.3.2 Correction for cover
In the above expressions it has been assumed that h – d § 0,1h and also that a problem of pure
bending is being analysed and therefore, hcr = 0.5h, kc = 0.4 and k’=1. Furthermore, the tensile
strength of concrete has been assumed to be 2.9 N/mm
2
(2.5 in case of EC2). If different values are
assumed for these parameters, then the value obtained from the tables must be corrected by the
following factor:
Part of cross section compressed:
( ) ( )
ct,eff ct,eff c cr c cr
f f k h k h 0,1h k' 1
=
2, 9 0, 4 0, 5h h d 1 2,9 2 h d 1
(7.29)
Cross section in tension:
( ) ( )
ct,eff ct,eff c cr c cr
f f k h k h 0,1h 1 1
=
2, 9 0, 4 0,5h h d k' 2,9 2 h d 2
(7.30)
It can therefore be written that:
( )
 
ct,eff * c cr
s s
f k h
=
2,9 2 h d
part of section compressed (7.31a)
( )
 
ct,eff * c cr
s s
f k h
=
2,9 4 h d
all section in tension (7.31b)
C7.3.4. Formula for crack width
7.3.4.1 Introduction
The formula proposed for crack width is a mixture of EC2 and MC90.
In order to evaluate the accuracy of this formula, it has been tested against an experimental data
base including results from the researchers Rehm & Rüsch [1113], Krips [9] , Falkner [6] , Elighausen
[5] , Hartl [7] , Beeby [1,2] and Jaccoud [8].
The data base has been drafted specifically for this document and is detailed in appendix A. The
criteria for the selection of the experimental results are clearly explained below. It is the intention of
the authors to avoid ambiguity and provide a self explaining instrument which can be used by other
researchers in the future so that the work carried out here need not be repeated. A detailed
presentation of this data base is therefore given.
After a review of the proposed model, the results of the comparison between model and experimental
data are presented. The performance of the new prEN formula is also compared to that of MC90 and
EC2. The results show small differences between the 3 models.
The analysis, however, shows that for all 3 models, the error margin grows as the crack width grows
and that all models tend to underestimate the crack width when it is large. This fact is unfortunate
since the control of cracking in normal structures is most important when cracks are large. This fact
suggests that in future editions of EC2, some correction might be needed to allow for this situation.
7.3.4.2 Proposed formulation
According to the prEN proposal, the design crack width can be determined using the following
expression:
wk = srm (csm – ccm)
where
wk design crack width
srmax maximum crack spacing
İsm mean strain in the reinforcement, under the relevant combination of loads, taking into account
the effects of tension stiffening, etc.
İcm mean strain in concrete between cracks
The strain difference (csm  ccm) may be calculated from the expression:
( )
( ) o
c c c >
ctm
s t eff
s eff s
sm cm t sr
s s s
f
ı  k 1+ ȡ
ı ȡ ı
 =  k = 0, 6
E E E
(7.32)
where
ıs stress in the tension reinforcement assuming a cracked section.
csr
( ) o
c ~
ctm
eff
sr eff
sr
s s
f
1 ȡ
ı ȡ
=
E E
This is a simplification which is exact for pure tension but not for
bending. However, this simplification makes it easier to apply the model in practical cases
and does not imply any significant loss of accuracy as is shown below.
oe ratio Es/Ec
µp,eff
s
c
A
=
A
Ac,eff effective tension area. Ac,eff is the area of concrete surrounding the tension reinforcement of
depth, hc,ef , where hc,ef is the lesser of 2,5(hd), (hx)/3 or h/2 (see figure).
Guide to EC2 Section 7
Page 78 Table of contents
kt factor dependent on the duration of the load
kt = 0,6 for short term loading
kt = 0,4 for long term loading
a) Beam b) Slab
A  level of steel centroid B  effective tension area
B  effective tension area
c) Member in tension
B  effective tension area for upper surface
C  effective tension area for lower surface
Figure 7.9 Definition of the effective tension area
In situations, where bonded reinforcement is fixed at reasonably close spacing within the tension zone
(spacing 5(c+ĳ/2), the maximum final crack spacing can be calculated from the expression:

rmax 1 2
eff
s = 3, 4c + 0, 425k k
ȡ
(7.33)
where
 bar diameter.
c cover to the reinforcement
k1 coefficient which takes account of the bond properties of the bonded reinforcement;
k1 = 0.8 for high bond bars
k1= 1.6 for bars with an effectively plain surface
k2 coefficient which takes account of the distribution of strain;
k2 = 0.5 for bending
k2= 1.0 for pure tension
For cases of eccentric tension or for local areas, intermediate values of k2 should be used
which can be calculated from the relation:
( ) c c
c
1 2
2
1
+
k =
2
(7.34)
Where İ1 is the greater and İ2 is the lesser tensile strain at the boundaries of the section
considered, assessed on the basis of a cracked section.
As can be seen, with regard to crack spacing, the cover, c, is introduced explicitly into the expression
of the crack width as suggested by A. Beeby. This suggestion is backed up by the following figure,
taken from reference [1], which clearly shows the dependence of the crack spacing on this parameter.
In EC2, c is implicitly taken as 25 mm.
Guide to EC2 Section 7
Page 79 Table of contents
Fig.7.10 Influence of cover on the transfer length acccording to Beeby [1]
Also, on a formal level, the formula of prEN gives srmax instead of srm.
srmax is obtained as 1.7 times srm. Thus,
1 2
p,eff
1, 7 2 25+ 0,25k k
ȡ

(7.35)
of EC2 becomes

1 2
p,eff
3, 4c + 0, 425k k
ȡ
(7.36)
of prEN.
In the case in which 2 different bar diameters are used, an equivalent diameter eq has to be
determined in order to apply the above formulation.
In EC2 [4] it is recommended to use the average diameter m.
MC90[3] and prEN [10] suggest for this case the equivalent diameter eq, although MC90 provides no
definition of ĳeq. The definition of eq depends on the definition of ȡeff. To show the difference, first m
will be applied to the equilibrium equation (7.33) to derive eq and then eq will be derived by
considering 2 diameters in the determination of the steel area (7.35).
ctm ctm t
f A = n l t t  (7.37)
(7.33) changes in case of the use of 2 different diameters into (7.34).
( ) tt  
ctm ctm 1 2 t
f A = n + n l (7.38)
Hence:
( ) tt  
ctm ctm
t
1 2
f A
l =
n + n
(7.39)
since srmax = 2lt and srmax = /3.6ȡ and, according to MC90[5]
t
ctm
f 1
=
1, 80
, it can be written:
( )
2

t tt  
ct,eff eq ctm ct,eff
rmax
eq 1 1 2 2
A 2f A
s = = 2
n + n
1, 80×4
4
(7.40)
From (7.36) can be seen that the equivalent diameter is:
( )  
 
1 1 2 2
eq m
1 2
n + n
= =
n + n
(7.41)
This is correct when the steel section in
s
ct,eff
A
ȡ =
A
is defined with ( )
t
2
eq
s 1 2
A = n + n
4
If this is not the case, and the steel area is instead described by
t t
2 2
1 2
s 1 2
A = n + n
4 4
then,
( )
t
 
2 2
1 1 2 2
c,eff
n + n
4
A =
ȡ
and therefore
( )
( )
 
t  
2 2
1 1 2 2
ctm
rmax
1 1 2 2
n + n
f
s = 2
ȡ n + n
(7.42)
From (1.38) can be derived that the equivalent diameter is:
( )
( )
 

 
2 2
1 1 2 2
eq
1 1 2 2
n + n
=
n + n
(7.43)
It can be seen that EC2 and MC90 could be misleading and thus should be clarified, since the code
user will naturally define ȡeff as As/Ac,eff. For this reason the definition of eq is made explicit in prEN.
Guide to EC2 Section 7
Page 710 Table of contents
7.3.4.3 Presentation of data base
As stated above, the data base includes data from different researchers. The data has been selected
among the different tests available following a few simple rules:
 The materials used in the tests need to be similar to the materials used today in the building
of structures. This rule leads to discard tests which use low bond rebars, or concrete qualities
less than 20 N/mm
2
and steel qualities of less than 400 N/mm
2
.
 The stress range should be a serviceability range. For this purpose, only results within a
stress range in steel from 150 to 350 N/mm
2
were considered for tests involving direct
actions. For indirect actions, steel stresses up to the yielding stress of steel were considered,
since theses test can be representative of walls subject to shrinkage and temperature.
 For the determining of the crack spacing the number of cracks present at the last phase of the
test is always considered since it is the closest to stabilized cracking, which is the crack
spacing given in equation (7.32).
7.3.4.4 Analysis of the experimental data
This paragraph includes the analysis of the experimental data as well as the comparison between the
values obtained from the experimental data base and the values obtained from the theoretical models.
The comparison includes not only the prEN model, but also those of EC2 and MC90, in order to
verify a satisfactory performance of the new proposal.
Experimental data generally includes the values for the mean crack width and, in most cases, also for
the maximum crack width.
The models are compared on the basis of the mean crack width because it is difficult to determine the
experimental characteristic crack width.
However, since in many of the experimental results the maximum crack width is also available, first an
analysis of the distribution function of the maximum crack width is made, since this analysis provides
some interesting conclusions. For this the results of Rehm & Rüsch [1113] are used.
7.3.4.4.1 Probability function of the maximum crack width
In this paragraph a first analysis of the probability distribution of the maximum crack width is made.
Figure 11 shows the crack width distribution function for all selected members within the service
stress range. The distribution function of the maximum crack widths shows clearly a tendency to a
normal distribution. Equation (7.44) shows the mathematical equation for the normal distribution
function. This expression is also plotted in Figure 7.11 showing very close agreement with the
experimental data.
( )
( )
 
µ


\ .
µ
t
2
2
x

2ı 1
f x, , ı = e
2 ı
(7.44)
Fig.7.11. Maximum experimental crack width distribution
Having proven that it is reasonable to assume that the probability distribution of the maximum crack
width can be assumed is a normal distribution, the distribution and the density functions were
evaluated for each stress level within the service spectrum: 200 N/mm
2
, 250 N/mm
2
, 300 N/mm
2
and
350 N/mm
2
.
These functions are displayed in Figure 7.12 (distribution function) and Figure 7.13 (density function).
The division of the results into different stress levels shows the evolution of the maximum crack width
as a function of the stress level. Figure 7.12 shows the distribution of the crack width for the different
stress levels. It can be seen that for a stress level of 200 N/mm
2
there is a probability of 95% that a
maximum crack width smaller than 0.3 mm occurs. This is consistent with the available experience. It
can be stated that for normal durability conditions, stress levels under 250 N/mm
2
will not pose any
Guide to EC2 Section 7
Page 711 Table of contents
serious cracking problems
Figure 7.12. Normal distribution of the maximum crack width for different stress levels
The density curves show the concentration of the crack opening. It can be seen in Figure 7.13, which
crack opening is typical for a certain stress level.
Fig.7.13. Density curves of the maximum crack width for different stress levels
This shows, that in the given data, a maximum crack width of 0.19 mm is typical for a steel stress of
200 N/mm
2
, 0.25 mm for 250 N/mm
2
and so on.
7.3.4.5 Comparison of the standards
In this paragraph a direct comparison between the formulae according to [3,4,10] is presented. This
comparison shows the performance of the formulae, not only against each other, but also against the
test results. Since prEN and MC90 provide the characteristic crack width, the mean crack width has
been estimated by dividing the value given by these codes by 1.7 and 1.5 respectively, since these
are the values they assume.
In Figure 7.14, all the data obtained by the evaluation of the experimental results and all the data
obtained by processing the corresponding input results according to the standards, is displayed.
Figure 14 Comparison testcalc., acc. to EC2, MC90 and PrEN
Guide to EC2 Section 7
Page 712 Table of contents
All 3 formulae perform quite closely to each other. This is shown by the corresponding trend lines.
A better view of this information can be obtained by plotting the error of the estimation instead of the
crack value. The error is defined as:
wm – wm,exp (7.45)
This result is given in the following graph and table:
Figure 7.15 Error in crack width prediction
Table 7.4. Errors in crack width prediction
The above table shows the mean value of the error, its standard deviation and the 95% confidence
intervals (assuming a normal distribution for the error) For example the mean error according to EC2
is werror = 0.005 mm (overestimation). With a 95% probability the underestimation will be less than
0.12 mm and the overestimation less than of 0.13 mm.
A normal distribution was assumed based on the following figure which shows the experimental error
compared to normal distribution having approximately the same mean value and standard deviation
(the theoretical value of the curve corresponds to EC2, but the values of the 3 models are fairly close).
It can be seen that good agreement is found and that the error can effectively be assumed have a
normal distribution.
Figure 7.16 Distribution function of the error in the estimation of the mean crack width
The diagram below shows the density function of the errors of each model.
Guide to EC2 Section 7
Page 713 Table of contents
Figure 7.17 Density function of the error in the estimation of the mean crack width
7.3.4.6 Conclusion
With the experimental data gathered, and the selection criteria given above, the crack formulae
[3,4,10] proposed for the crack prediction for the 3 different models discussed above was evaluated.
During the process of gathering and computing all the result data, three things became manifest:
 Within a steel stress limit of 250MPa the crack width does not have to be necessarily
checked. This was confirmed by the statistical evaluation, which demonstrated that below 250
N/mm
2
the anticipated maximum crack opening will not exceed a value of 0.4 mm with a
plausibility of 95%.
 The error of the formulae in the evaluation of the crack width [3,4,10] increases for larger
crack openings. Also all models tend to underestimate the crack width when it is large. This is
a critical point, because especially for higher steel stresses, the crack opening is expected to
be critically larger and the prediction more important than for small cracks under lower steel
stress. It might be interesting for future proposals of the EC2 crack prediction formula to
provide an adjustment to compensate this tendency.
 Nevertheless it could be verified that the existing models, EC2 [4] and MC90 [3] and the PrEN
[10], provide acceptable predictions. The mean value of the observed error is in all cases
close to zero. Also the standard deviation of this error is relatively small (0.063 to 0.076 mm).
With all this information it can be said, that the PrEN has a good performance range and will be an
adequate substitute for the existing EC2 formula.
87.3.4.7.1 Exact derivation
The EC2 cracking opening formula is:
s,cr s
k 3 1 2 4
s s s
w 1 k c k k k
E
o ( ( o 
= ÷ + ì
( (
o µ
¸ ¸ ¸ ¸
(7.46)
Assuming the prescribed values k3=3.4, k4=0.425 and considering the bending case (k2=0.5) with high
bond reinforcement (k1=0.8), it results
s,cr s
k
s s s
w 1 3.4 c 0.17
E
o ( ( o 
= ÷ + ì
( (
o µ
¸ ¸ ¸ ¸
(7.47)
which can be used as a design formula. In particular, for given b, h, d, d’, b, and fixed M, we want to
deduce the steel reinforcement amount As and its design tension os in order to have a crack width wk
lower than the fixed value k
w
. The nondimensional depth of the neutral axis is
( ) ( )
2
e s e s
1
1 ' 0
2
÷ ç ÷o µ + ç +o µ o + o =
(7.48)
The steel tension is
( )
( ) ( )
e ctm t
s
2 2
3
s
f k
2 3n '
o v o ÷ç
o =
( (
µ o ÷ç + o ÷ç + ç
¸ ¸ ¸ ¸
(7.49)
Assuming
2 0
cr
t ctm
M M
b h M
k f
6
v = =
we get (7.50)
( )
2
s
e
2 1 '
ç
µ =
o ÷ + ç + o +o (
¸ ¸
(7.51)
Guide to EC2 Section 7
Page 714 Table of contents
( )
( ) ( ) ( )
3 2
e
2 2
2 p 3 p
' 1 '
o v o ÷ç ÷ ç ç
=
o +o ÷ + ç o ÷ç + o ÷ç
where p=os/(ktfctm) (7.52)
When k
w w =
, after some calculations we deduce
0
k
s
s
w
p n
3.4 c 0.17
ì
= + +
 ì
µ
+
µ
(7.53)
setting
s k
0k
t ctm
E w
w
k f
=
Combining the above equations it results
( )
( ) ( )
( )
( )
( )
( ) ( )
0 2
e e k
e 2 2 2 2
e
2 3
2 2
2 w
' 1
3.4 c 0.34 ' 1
'
3 2
' 1
'
(
o v o÷ç o ì ç
= + o+o ÷ + ç +o × (
(
¸ ¸
ç + o ì o+o ÷ + ç ç ( (
o÷ç + o ÷ç ( ¸ ¸ ¸ ¸
¸ ¸
(
ç ç
(
× +
( o+o ÷ + ç ( (
o÷ç + o ÷ç
¸ ¸
¸ ¸ ¸ ¸
(7.54)
which numerically solved, gives the neutral axis position from which the reinforcement tension and its
amount can be determined. If it is not the case, it is necessary to set in the Ȝ = 2.5·(1o) and then re
evaluating ç, being the value Ȝ =0.5 practically impossible for bending problems.
The procedure, aimed to the determination of the reinforcement amount and its tension corresponding
to fixed crack width values and stress level, requires to assume the value of the bars diameter .
Alternatively, it is possible to set the tensional level os, for example equal to the limit one, and to
evaluate the corresponding reinforcement amount µs and the maximal bar diameter. In this case, as
the parameter p is defined, the neutral axis is calculated from (7.52), µs from (7.50) and the maximal
diameter can be derived solving with respect to , which gives
s s ok
max
e s
5.88 w
2c
(p )
( µ µ
 = ÷
(
ì ÷o µ ÷ì
¸ ¸
(7.55)
7.3.4.7.2 Approximated derivation
The procedure discussed above is quite laborious as it requires iteration. An alternative procedure,
easier to be applied, is based on the assumption of a lever arm h0 = 0.9d constant and independent
from ç.Therefore, osAs0.9d=M and
µs=0.185v/(po) (7.56)
The general formula for k
w w =
gives
s e s
k
s s s
w 1 1 3.4c 0.17
E p
(   o o µ ì  ì
 
= ÷ + +
  (
µ ì µ \ .
¸ ¸ \ .
(7.57)
For further simplification of the problem, assuming o=0.9, therefore Ȝ = 0.243 and by definition
*
1.18
1 1
0.185
v v
v = =
o ì
÷ ÷
v v
1
c
u =

0k
2
w
u =

(7.58)
the following equation is obtained
 
2
e
1 e 1 2
p 5 * 3.4u 0.20 p * 17 u 5u 0
*
o
(
v
+ v ÷ ÷v o + =
v v (
¸ ¸
(7.59)
This formula is easy to solve and leads to the desired values µs and os.
In this case too, for a given value of os, solving for  we obtain
* *
e ok
max *
2
e
17c( p ) 5 w
p p
( v ÷o v ÷ v
¸ ¸
 =
v
o ÷
v
(7.60)
that defines the maximal bar diameter, which satisfies the cracking limit state corresponding to a fixed
value of the tension in the steel
Guide to EC2 Section 7
Page 715 Table of contents
7.4 Deflection control
7.4.1 General considerations
7.4.2 Cases where
calculations may be omitted
7.4.3 Checking deflections by
calculation
C7.4 Discussion of the general method followed for deflection calculation
7.4.1 Instantaneous deflections
Instantaneous deflections are computed by applying the total load on a structure in which there is a
reduction of the stiffness. The law to calculate the reduction of stiffness is deduced from equation 7.8
of prEN 19921:
( ) ( )
   
ç ç ç + ç
 
\ . \ .
II I e I I
1 M 1 1 M M
= = + 1 = 1
r EI r r EI EI
(7.61)
From this equation, the following relationship is obtained:
( ) ç ç
I II
e
I II
I I
I =
I + 1 I
(7.62)
The coefficient applied to the stiffness of each section is obtained as:
e
I
I
coef =
I
(7.63)
7.4.2 Longterm deflections
7.4.2.1 Assumed Load History
The load history influences the value of the deflections. In this study a realistic load history has been
taken into account. A typical load history for buildings could be:
• Application of self weight at 10 days
• Application of the remaining dead load at 60 days
• Application of quasipermanent load at 365 days.
7.4.2.2 Deflections due to creep
Complex Loading History The interpretation of prEN 19921 regarding deflections due to creep is not
clear when the loadhistory is complex.
As described in paragraph 7.4.2.1, the assumed load history involves 3 dates for the application of the
loads:
• Application of self weight of the structure, g1 at time t1
• Application of remaining dead load g2 at time t2
• Application of quasipermanent live load ¢02q at t3
prEN 19921 proposes to take into account the creep of concrete by using an effective modulus for
concrete:
¢
c,eff
1
E =
1+
(7.64)
The question then arises about which time should be used to compute the creep coefficient.
To solve this problem the following procedure is proposed:
• Compute fg1,(t1, ç1) with ĳ(t,t1) and ç1.
• Compute fg1+g2(t2, ç2) with ĳ(t,t2) and ç2.
• Compute fg1+g2+q(t3, ç3) with ĳ(t,t3) and ç3.
• Compute fq1(t2, ç1) with ĳ(t,t2) and ç1.
• Compute fg1+g2(t3, ç2) with ĳ(t,t3) and ç2
• Compute the total deformation by using the following expression:
fg1 +g2+q,·
= fg1,(t1,ç1) + fg1+g2(t2,ç2)  fq1(t2,ç1) + fg1+g2+q(t3,ç3)  fg1+g2(t3, ç2) (7.65)
This complicated and timeconsuming procedure is necessary due to progressive cracking of cross
sections. This expression takes into account, for example, that part of the deflection due to g1 occurs
at time t2 due to the reduction of stiffness produced by the application of load g2. The creep of this
extra deflection must therefore be referred to time t2. This is what is achieved by the above
expression.
In case a construction live load equivalent to the value of g1+g2+ȥ01q is assumed to be applied at time
t1, the above expression is greatly simplified, since no reduction of stiffness occurs after the
application of g1+construction load:
fg1 +g2+q,·
= fg1,(t1,ç3) + fg1+g2(t2,ç3)  fq1(t2,ç3) + fg1+g2+q(t3,ç3)  fg1+g2(t3, ç3) = fg1,(t1,ç3) + fg2+q(t2,ç3) + fq(t3,ç3)
(7.66)
7.4.2.3 Deflections due to shrinkage
Deflections due to shrinkage are computed for a stiffness corresponding to the quasipermanent load
condition, taking into account an effective modulus with a creep coefficient corresponding to the start
of development of shrinkage (i.e. end of curing), ¢(t,ts).
Also, II and III are calculated using Ec,eff.
7.4.3 Parametric study of slenderness limit
7.4.3.1 Introduction
ENV 199211:1991 in table 4.14 and MC90 in table 7.5.2 provide slenderness limits for lightly
reinforced and heavily reinforced concrete elements. In both cases, for a simply supported beam the
Guide to EC2 Section 7
Page 716 Table of contents
corresponding values are, respectively, L/d=18 and L/d=25. These values are given for a steel yield
stress of 400 N/mm
2
, and are inversely proportional to the steel grade. This means, that the
equivalent values for a 500 N/mm
2
steel would be L/d=14 and L/d=20.
There have been complaints in the sense that this table is too conservative, or too general. The
parametric study described in the following sections, considers a large range of variables affecting the
deformation of concrete structures, in order to quantify their influence and study the possibility of
including them in the calculation of the slenderness limit. The present proposal for section 7.4 of prEN
19921, is based on this study.
The parametric study, which has been carried out according to the procedure described above,
considers the influence on the slenderness limit of the following parameters:
• Complex load history. The influence of the values of t1, t2 and t3 on the slenderness limit has been
studied.
• Control of total deflections vs. control of deflection which produced cracking of partitions (referred
to in this document as active deflection). Slenderness limits are calculated by limiting the total
deflection to L/250. It has been investigated whether the limitation of the deflection producing
cracking of partitions to L/500 can be more restrictive.
• Influence of relative humidity. The relative humidity affects long term deflections through creep
and shrinkage. The influence of this parameter on the slenderness limit has been studied for
relative humidity varying from 50 to 80%.
• Real reinforcement vs. required reinforcement. The effect of considering a 5% to 10% increase in
the real reinforcement with respect to the required reinforcement determined from U.L.S. analysis
has been studied in order to take into account the roundoff in detailing.
• Distribution of reinforcement. Reinforcement in real beams is not constant. The influence of the
real distribution of reinforcement on the slenderness limit has been studied.
• Concrete grade
• Percentage of self weight (g1), additional deal load (flooring and partitions g2) and quasi
permanent live load (ȥ02q) with respect to the total load (qtot). According to Spanish practice,
typical values for these relations could be:
 For oneway slabs,
¢
1
tot
2
tot
02
tot
G
= 0, 45
Q
G
= 0, 30
Q
Q
= 0, 30 0, 25 = 0,075
Q
 For flat slabs,
¢
1
tot
2
tot
02
tot
G
= 0,60
Q
G
= 0,20
Q
Q
= 0,30 0, 20 = 0,06
Q
For this study, as for prEN 19921, the slenderness limit is defined as the relationship between the
span and the effective depth L/d.
The slenderness limit curves which are presented in the following paragraphs, are given for different
reinforcement ratios. The reinforcement ratio, ȡ, is defined as the ratio of tensile reinforcement As, to
effective cross section bd.
7.4.3.2 Assumptions for parametric study
The parametric study which follows has been carried out for a simply supported beam with a cross
section of 100 x 30 cm
2
. The cover has been assumed as 1/10 of the total depth.
The reference values for which the study is formulated are the following:
 Relative humidity of 70%
 Load history: t1/t2/t3 =10/60/365 days
 Permanent load vs. live load: g1 = 45%qtot, g2 = 30%qtot and q = 25% of qtot.
 Quasi permanent live load is 30% of characteristic live load
 Tensile and compressive reinforcements are those strictly needed for ULS.
 Concrete Strength: 30 N/mm
2
 Steel Yield Stress: 500 N/mm
2
 Distribution of reinforcement is considered constant over the beam length.
For each part of the study, one of the above parameters is varied while the others remain constant.
Guide to EC2 Section 7
Page 717 Table of contents
7.4.3.3 Method for determining the slenderness ratio
In order to determine the slenderness ratio, the following steps were taken for each reinforcement
ratio:
 A certain span length, L, is assumed.
 Calculation of the ultimate bending moment (MULS). When compression reinforcement is needed in
order to yield the tensile reinforcement, compressive reinforcement is provided and taken into account
in the calculation of deflections.
 The ultimate load, qULS, is determined from the MULS, assuming a simply supported beam:
ULS
ULS 2
M
q = 8
L
 The total service load (qtot) is determined from qULS, and the assumed ratios for g1/qtot, g2/qtot and
q/qtot, according to:
qULS = 1.35ā (g1 + g2) + 1.5ā q = 1.35ā (0.45 + 0.30)ā qtot + 1.5ā 0.25ā qtot = 1.38ā qtot ĺ qtot = 0.72ā qULS
 The values of g1, g2 and q are determined from the above ratios.
 The deflection is computed. According to the general method of prEN 19921. If the deflection
obtained is not L/250 for total deflection or L/500 for active deflection, the procedure is repeated until
convergence is achieved.
7.4.3.4 Influence of the dimensions of the cross section used
The cross section assumed for the parametric study is, as stated above, a rectangular cross section
of b x h = 100 x 30 cm
2
. In order to insure that the particular dimensions of the cross section are not
important, the slenderness limit for different reinforcement ratios has also been determined for the
cross section used by Beeby in [2], all other parameters being those taken as reference (see section
3.2).
This cross section is rectangular of dimensions b x h = 30 x 50 cm
2
.
Figure shows the comparison between both rectangular cross sections.
As can be seen no significant difference can be observed.
Figure 7.18. Slenderness ratio for two rectangular cross section of different dimensions
7.4.3.5 Influence of load history
The load history used for the parametric study is described above in 2.2.1. It is assumed that the
construction live load is applied at the same time as the self weight so that the section is fully cracked
from the beginning. This provides an upper bound estimation for total deflection (not so for active
deflection).
The reference load history is: t1/t2/t3=10/60/365 days
Two other load histories are considered: t1/t2/t3=7/14/365 days and t1/t2/t3=28/90/365 days.
The calculation of the slenderness limit has been carried out in this case for two steel ratios: 0.5% and
1.5%.
The results are shown in Table 7.5. As can be seen, the influence of the load history is very limited.
This suggests that simplifications are possible. One such simplification, consisting in considering a
single time of loading together with an equivalent creep coefficient is described in detail in section
7.4.4.1.
Table 7.5. Influence of load history on the slenderness limit.
((L/d)(10,60,365) (L/d)(t1,t2,t3))/ (L/d)(10,60,365)
Guide to EC2 Section 7
Page 718 Table of contents
7.4.3.6 Influence of additional reinforcement
The slenderness ratio is referred to the deflection occurring in case the strict reinforcement is placed
in the beam. However, it is normal for structures to have extra reinforcement due to rounding off, on
the safe side, of the required values.
The influence of this factor has been studied by comparing the deflection of a beam with different
required reinforcement ratios in case an extra 5% or 10% reinforcement is provided in tension. The
results are given in Fig. 7.19.
Figure 7.19. Influence of round–off extra reinforcement in the deflection of a simply supported beam
7.4.3.7 Influence of distribution of reinforcement
Distribution of reinforcement in beams is not constant in practice.
Normally, minimum reinforcement is placed near the points of zero bending moment (base
reinforcement) and a supplementary reinforcement is placed in the areas near the maximum bending
moment. The influence of the length of the additional reinforcement has been studied for two
reinforcement ratios (0.5% and 1.5%) and assuming a length (ls) of the supplementary reinforcement
of 60, 80, 90 and 100% of the span. Outside this length minimum reinforcement was considered.
The results are given in Table 7.6 by comparing the deflection for ls=100% to the deflection for the
different values of ls. It can be seen that for low values of the reinforcement ratio and no influence is
detected. The difference becomes significant only for high reinforcement ratios and small length of
additional reinforcement.
Table 7.6. Influence of the length of supplementary reinforcement on the deflection of
a simply supported beam.
7.4.3.8 Active deflection vs. total deflection
Table 4.14 of ENV199211:1991 as well as table 7.4 of prEN 19921, has been determined by
limiting total deflections. However, as can be seen from figure Fig. 7.20, the limit of active deflection to
a maximum value of L/500 is a more strict condition. Further consideration should be given to this
fact.
Figure 7.20 – Slenderness Limit for Total and Active Deflection
Guide to EC2 Section 7
Page 719 Table of contents
7.4.3.9 Influence of relative humidity
The influence of relative humidity on the slenderness limits was studied for two different steel ratios: ȡ
= 0.5% and ȡ = 1.5%. The slenderness limit was determined for these two steel ratios and relative
humidity of 50, 60, 70 and 80%. The results are given in Fig. 7.21.
Figure 7.21. Influence of relative humidity on Slenderness Limits for Total and Active Deflections (referred to 70% RH)
From Fig. 7.21 it can be seen that the influence of relative humidity on the slenderness ratio is
important with a variation of ±15% for the active deflection and ±10% for total deflection from the
reference value of 70% of relative humidity.
The figure also shows that the influence of relative humidity does not significantly depend on the steel
reinforcement ratio.
7.4.3.10 Influence of the percentage of self weight, superimposed dead load and live load with
respect to total load.
For this study, different load distributions have been assumed, including the assumption that the quasi
permanent service load, qcp, can be taken as 0.5 times the ultimate limit state (ULS) load.
The ULS load is given by the following expression:
qULS = 1.35ā (g1 + g2) + 1.5ā q
As explained in section 7.4.3.3, the ultimate load qULS is known from the value of the reinforcement. If
it is assumed that g1 § 1.5ā g2, then by fixing different values for the quasi permanent service load
(qcp)/ultimate load (qULS) ratio, g1, g2 and q may be determined for each ratio. This provides, in the
end, significantly different values for the permanent to live load ratios for the quasi permanent
combination. The lower the value of the permanent load, the lower will be the load considered for the
verification of deflections, since only 30% of the live load is considered quasi permanent.
In Table 7.7, the different ratios of qcp/qULS considered and the resulting ratios of g1/qtot, g2/qtot and
q/qtot are given.
Table 7.7. Total service load/Ultimate load ratios and corresponding values of
self weight, superimposed dead load and live load to total service load ratios
The comparison of the results of the slenderness limits for the different quasi permanent service load
to ultimate load ratios is plotted in Fig. 7.22. As can be seen the difference is important. It is also
interesting to note that the values of l/d = 20 for 0.5% reinforcement ratio and l/d=14 for 15%
reinforcement ratio, are obtained approximately for a ratio of qcp/qULS=0.5.
Guide to EC2 Section 7
Page 720 Table of contents
Comparison Different load distribution (g1/qtot), (g2/qtot), (q/qtot)
Figure 7.22. Slenderness limits for total deflection and different total service load to ultimate load ratios
7.4.3.11 Influence of the concrete grade
The influence of the concrete grade on the slenderness limit has been studied in a very detailed
manner for many values of the reinforcement ratio. The concrete grades considered are: C30,
C40,C60 and C100.
For this comparison, the relationship between total service load and ultimate load has been taken as
51%. This value, is different from the reference value, but, as explained in section 7.4.3.10, agrees
well with the values of prEN 19921 table 4.14 which are also in accordance with the values of MC90
and ENV199211:1991.
Fig. 7.23 shows the results of this comparison. It can be seen that there is a very significant influence
of the concrete grade on the slenderness limit. These curves, also show very well the behaviour for
very low steel ratios. There is a clear discontinuity in the curves which separates the point in which
cracking does not occur from the point in which the bending moment is greater than the cracking
moment. The curves also show that for low steel ratios, much higher slenderness limits can be
established. These curves form the basis for the formula proposed for the slenderness limit in prEN
19921, which is dependent on the concrete grade, and provides a continuous estimation of the
slenderness limit as a function of the reinforcement ratio.
Figure 7.23. Slenderness ratio as a function of the concrete grade
7.4.3.12 Influence of the tensile strength of concrete (fctm vs. fctm,fl)
The final version of prENV, allows the use of the flexural tensile strength of concrete instead of the
mean tensile strength for the calculation of deflections. This topic was subject of much controversy
and it is therefore interesting to test the differences to which this may lead. For this purpose, the
slenderness limits were redetermined with the assumption that the flexural tensile stress of concrete
could be used. The results are shown in Fig. 7.24.
Guide to EC2 Section 7
Page 721 Table of contents
Figure 7.24. Slenderness Limit using fctm and fctm,fl
As can be seen there is an important difference only for very small reinforcement ratios. For
reinforcement ratios larger than 0.6%, the difference is negligible.
7.4.4 Simplified formulae
7.4.4.1 Simplified formula of EC2
The application of the general method of prEN 19921 is very tedious and time consuming, since
calculations have to be made for many sections. As an alternative to this procedure a simplified
method consisting in calculating an equivalent moment of inertia as explained in section 2.1 for the
centre span of the beam only and assuming this value for the whole beam is also recommended in
the present draft for prEN 19921. This procedure in on the safe side, since cross sections near the
point of zero bending moment will not crack.
Additionally, in this comparison and in order to keep the simplified formula simple, an equivalent creep
coefficient is used. This creep coefficient which allows to take into account the load history in a
simplified manner is defined as:
( ) ( ) ( )
1 1 2 2 02 3
eq
1 2 02
g t,t +g t,t + q t,t
=
g +g + q
  ¢ 

¢
(7.67)
The general and simplified procedure are compared for the reference values described in section
7.4.3.2, in Fig.7.25. The comparison is made in terms of the differences in the value of the
slenderness limit for different reinforcement ratios. As can be seen, both procedures yield practically
equivalent for steel ratios greater than 1%. For a 0.5% reinforcement ratio, the error is about 5% on
the safe side. Fig. 7.26 shows a comparison of the differences obtained by plotting the ratio of
(L/d)simplified to (L/d)general as a function of the steel ratio. In this case a maximum difference of 20% in
obtained for the lowest reinforcement ratio considered.
Figure 7.25. Comparison of general and simplified methods of prEN 19921 proposal
Guide to EC2 Section 7
Page 722 Table of contents
Figure 7.26. Comparison of general and simplified methods of prEN 19921 proposal –
Evaluation of the relative error
7.4.5 Formula proposed for slenderness limit
In order to take into account the influence of the concrete grade on deflections, which is important,
especially for low steel ratios, and in order to provide designers with a continuous relationship
between the slenderness ratio and the reinforcement ratio, which is especially helpful for slab
elements, the following expressions have been proposed.
3
0 ck
ȡ = f 10 (7.68)
3
2
0 0
ck ck 0
ȡ ȡ l
=k 11+1.5 f +3.2 f 1 if ȡ<ȡ
d ȡ ȡ
(
 
(

(
\ .
¸ ¸
(7.69)
0 0
ck ck 0
ȡ ȡ l
=k 11+1.5 f +3.2 f if ȡ ȡ
d ȡȡ' ȡ
(  
> ( 

(
\ . ¸ ¸
(7.70)
These formulas are a mathematical approximation to Fig. 7.23. In Fig. 7.27, this approximation is
evaluated. Four plots are presented for the four concrete grades considered in section 7.4.3.11. As
can be seen the degree of coincidence between the formula and the calculated values of the
slenderness limits is very good, and mostly on the safe side.
Concrete Grade C30
Guide to EC2 Section 7
Page 723 Table of contents
Formula  Concrete Grade C45
Concrete Grade C60
Concrete Grade C100
Figure 7.27. Comparison of General procedure with simplified formula for slenderness limit
Guide to EC2 Section 7
Page 724 Table of contents
REFERENCES
Cracking
[1] A.W. Beeby, Calculation of Crack Width, PrEN 19921 (Final draft) Chapter 7.3.4
(2001).
[2] A.W. Beeby, R. Favre, M. Koprna, J.P. Jaccoud, Cracking and Deformations, CEB
Manual (1985).
[3] CEB, Draft of chapter 15, Model Code 1990, (1987).
[4] EC2, Planung von Stahlbeton und Spannbetontragwerken, Teil 11 (1996).
[5] R. Elighausen, R. Mallée, G. Rehm, Rissverhalten von Stahlbetonkörpern bei
Zugbeanspruchung, Untersuchungsbericht (1976).
[6] M. Falkner, Zur Frage der Rissbildung durch Eigen – und Zwangspannungen in Folge
Temperatur im Stahlbetonbau, Deutscher Ausschuss Für Stahlbeton Heft 208 (1969).
[7] G. Hartl, Die Arbeitslinie eingebetteter St¨¨ale bei Erst – und Kurzzeitbelastung,
Dissertation (1977).
[8] J.P. Jaccoud, H. Charif, Armature Minimale pour le Contrôle de la Fissuration,
Publication Nº 114 (1986).
[9] M. Krips, Rissenbreitenbeschränkung im Stahlbeton und Spannbeton, TH Damstadt
(1985).
[10] PrEN, 19921, Serviceabillity Limit States, Final Draft (2001).
[11] G. Rehm, H. Rüsch, Versuche mit Betonformstählen Teil I, Deutscher Ausschuss für
Stahlbeton Heft 140 (1963).
[12] G. Rehm, H. Rüsch, Versuche mit Betonformstählen Teil II, Deutscher Ausschuss für
Stahlbeton Heft 160 (1963).
[13] G. Rehm, H. Rüsch, Versuche mit Betonformstählen Teil III, Deutscher Ausschuss für
Stahlbeton Heft 165 (1964).
Deflection
[1] CEN. Eurocode 2: Design of concrete structures – Part 1: General rules and rules
for buildings. Ref. No.. prEN 19921:2000. 1st Draft.
[2] Beeby, A.W. 4.3.3 Deformation. fib Bulletin 2 – Structural concrete Textbook.
Volume 2. July 1999
[3] CEBFIP. Model Code 90. Bulletin d’Information Nº213/214. May 1993.
[4] CEN. Eurocode 2: Design of concrete structures – Part 1: General rules and rules
for buildings. Ref. No.. prEN 19921:1991.
Guide to EC2 Section 8
Page 81 Table of contents
SECTION 8 DETAILING OF
REINFORCEMENT AND
PRESTRESSING TENDONS 
GENERAL
8.1 General
8.2 Spacing of bars
8.3 Permissible mandrel
diameters for bent bars
8.4 Anchorage of longitudinal
reinforcement
8.5 Anchorage of links and
shear reinforcement
8.6 Anchorage by welded bars
8.7 Laps and mechanical
couplers
8.8 Additional rules for large
diameter bars
8.9 Bundled bars
8.10 Prestressing tendons
SECTION 8. DETAILING OF REINFORCEMENT AND PRESTRESSING TENDONS  GENERAL
See example 6.15
See example 6.15
See example 6.15
C8.10.2.
8.10.2.1. Background
The ENV rules for the anchorage of pretensioned tendons are found in 4.2.3.5.6. The basic value of
the transmission length for prestress at release (transfer) is given as a multiplier ȕb for the nominal
diameter, table 8.1.
Table 8.1. Basic value of factor ȕb for transmission length according to ENV, table 4.7
Two design values of the transmission length are defined, 0,8 and 1,2 times the basic value, of which
the more unfavourable one should be used depending on the design situation.
For anchorage at the ultimate limit state, the bond is assumed to be the same as at release of
prestress.
The ENV rules did not take into account the following parameters:
a) the initial prestress in the tendons
1
b) the way of release of prestress (gradual or sudden)
2
c) the difference in bond conditions with regard to the position of tendons etc.
d) the difference in bond conditions with regard to “pushin” and “pullout”
3
e) the difference between strands and indented wires.
The national comments on clause 4.2.3.5.6 were mainly editorial, but the rules needed a thorough
update based on MC90, which represents a more uptodate knowledge, and takes into account all
the important parameters.
The present rules in clause 8.10.2 of the EN are basically a different formulation of clause 6.9.11 in
MC90. There is one difference, however. MC90 gives two design values of the transmission length,
with a ratio 2,0 between the upper and lower value. The rules in 8.10.2 have been calibrated to give
the same mean value as MC90, but with a ratio 1,2/0,8 = 1,5 between the upper and lower values,
like in the ENV. Thus, the upper value will be o lower and the lower value somewhat higher according
to the EN, compared to MC90.
Among the ENV rules the alternative parabolic development of prestress, given in ENV 199213
should also be mentioned. This has not been explicitly included in the EN, since a linear development
is considered to be more realistic. However, an opening for “alternative buildup of prestress” is given,
without details.
8.10.2.2. Rules in MC90 and EN 199211
8.10.2.2.1 Summary of rules
The rules in MC90 and EN 199211 are summarized in table 8.2. They are described and
commented more in detail in the following clauses.
1
It is stated in the ENV that table 4.7 (table 1 above) is valid for the maximum allowable prestress, but nothing is said
about reducing the transmission length for lower values of prestress
2
A ”neutralized zone” for the case of sudden release is mentioned in 4.2.3.5.6 (5), but no value is given
3
”Pushin” refers to the situation at release of prestress, whereas ”pullout” refers to anchorage in ULS
Guide to EC2 Section 8
Page 82 Table of contents
Table 8.2. Summary of rules in MC90, 6.9.11, and EN 1992, 8.10.2
8.10.2.2.2 Rules in MC90
8.10.2.2.2.1 Basic anchorage length
t
sp pd
bp
bpd
A f
l =
f
where
Asp/t = /4 for circular tendons, (7/36) for 7wire strands
fpd = fptk/¸s design value of tensile strength of tendons
fbpd design value of bond strength, see below
8.10.2.2.2.2 Design value of bond strength
fbpd = qp1qp2fctd(t)
where
qp1 takes into account the type of prestressing tendon:
qp1 = 1,4 for indented and crimped wires and
qp1 = 1,2 for 7wire strands
qp2 takes into account the position of the tendon:
qp2 = 1,0 for good bond condition,
qp2 = 0,7 for all other cases.
fctd(t) is the lower design value of concrete tensile strength at time of release, or at 28 days for
verifications in ULS.
8.10.2.2.2.3 Transmission length
o
o o o
pi
bpt 8 9 10 bp
pd
l = l
f
where
o8 considers the way of release:
o8 = 1,0 for gradual and 1,25 for sudden release
o9 considers the action effect to be verified:
Guide to EC2 Section 8
Page 83 Table of contents
o9 = 1,0 for calculation of anchorage length when moment and shear capacity in ULS
is considered, and
o9 = 0,5 for verification of transverse stresses in the anchorage zone
o10 considers the influence of bond situation:
o10 = 0,5 for strands and
o10 = 0,7 for indented or crimped wires
ıpi is the steel stress just after release
8.10.2.2.2.4 Anchorage length in ULS
o o
pd pcs
bpd bpt bp
pd

l = l + l
f
where
ıpd tendon stress under design load (ıpd fpd)
ıpcs tendon stress due to prestress including all losses
8.10.2.2.3 Rules in EN 199211
The rules in 8.10.2 are summarized below and compared to those in MC90. The “basic anchorage
length” used as a separate parameter in 8.10.2 is instead incorporated directly in the expressions for
transmission length and anchorage length.
8.10.2.2.3.1 Transfer of prestress
8.10.2.2.3.1.1 Bond strength
The bond strength governing the transmission at release of prestress is
fbpt = qp1 q1 fctd(t)
where
qp1 takes into account the type of tendon and the bond situation at release (”pushin”)
= 2,7 for indented wires (“crimped” wires are hardly used anymore, therefore not incl.)
= 3,2 for 7wire strands
q1 = 1,0 for good bond conditions (see 8.4.1) or = 0,7 otherwise, unless good bond conditions can
be verified
fctd(t) = fctk,0,05(t) /¸C, design value of tensile strength at the time of release, related to the
compressive strength at the same time according to table 3.1
The factor q1 is the same as in 8.4.1 (and the same as Șp2 in MC90). However, a possibility to
assume “good bond conditions”, even if the criteria in 8.4.1 are not met, has been added for cases
where good conditions can be achieved by other means, and verified.
The factor Șp1 here includes two factors, which in MC90 are applied to the transmission length
instead, namely the factor o10 and the mean value of o9, (0,5 + 1,0)/2 = 0,75:
q
q q
o o
q
p1,MC90
p1,EN p1,EN
10 9,mean
p1,EN
1,4
= which gives = = 2,7 for wires
0,7×0,75
1,2
= = 3,2 for strands
0,5×0,75
Thus, fbpt includes the favourable effect of “pushin” at release. This is the “bond situation” to which
MC90 refers in the definition of o10: when the stress decreases at release there is transverse
expansion of the tendons, giving a kind of “wedge” effect. Furthermore, fbpt is here a mean value of
the bond strength, not a lower limit as in MC90 (the upper limit corresponds to o9 = 0,5 for the
transmission length). See 8.10.2.2.3.1.2 for upper and lower design values.
8.10.2.2.3.1.2 Transmission length
The basic value of the transmission length is
lpt = o1o2  ıpi/fbpt
where
o1 = 1,0 for gradual release, 1,25 for sudden release (same as o8 in MC90)
o2 = 0,25 for circular tendons, 0,19 for 7wire strands (cf. MC90: 7/36 § 0,19)
ıpi stress in tendon just after release (same as MC90)
Normally a short transmission length gives higher transverse stresses in the concrete at release
(spalling, splitting and bursting stresses in the terminology of MC90), whereas a long transmission
length is more critical for ULS with regard to shear, bending moment etc. Furthermore, there is an
uncertainty in the calculated value. Therefore, the more unfavourable of the following two values
should be used as a design value, depending on the design situation:
lpt1 = 0,8 lpt
lpt2 = 1,2 lpt
The factors 0,8 and 1,2 are the same as in the ENV, 4.2.3.5.6 (4). The corresponding factor in MC90
is o9 with the values 0,5 and 1,0. With the mean value of this factor included elsewhere, see
8.10.2.2.3.1, comparable values of o9 would be 0,67 and 1,33.
Guide to EC2 Section 8
Page 84 Table of contents
8.10.2.2.3.3 Anchorage in ULS
8.10.2.2.3.3.1 Bond strength
The bond strength for anchorage of stresses above prestress is
fbpd = qp2 q1 fctd
where
Șp2 takes into account the type of tendon and the bond situation at anchorage (”pullout”)
= 1,4 for indented wires
= 1,2 for 7wire strands
Ș1 same as above
fctd = fctk,0,05 / ¸C, design value of concrete tensile strength
Șp2 has the same values as Șp1 in MC90. and there is no favourable effect of “pushin”. Instead the
bond situation at anchorage is characterized by “pullout”, where the tendons “shrink” in the
transverse direction when the stress increases.
For fctd there is a limitation to the value valid for C55. The reason is that the linear relationship
between fbpd and fctd cannot be expected to be valid for higher concrete strengths. This is due to an
increasing brittleness, which results in a more uneven distribution of the bond stress. The average
bond strength will normally increase also above C55, although not in proportion to the tensile strength.
This may be taken into account, but requires a special verification.
8.10.2.2.3.3.2 Anchorage length
The anchorage length is based on the upper design value of the transmission length and the increase
of stress above the remaining prestress, with bond strength according to 8.10.2.2.3.3.1:
lbpd = lpt2 + o2  (ıpd  ıp) / fbpd
where
lpt2 the upper design value of the transmission length (see 8.10.2.2.3.2.2)
ıpd tendon stress to be anchored (same as in MC90)
ıp the remaining prestress (after all losses; same as in MC90)
8.10.2.3. Numerical comparisons
8.10.2.3.1 Transmission length
8.10.2.3.1.1 Effect of concrete strength
The transmission length according to EN, MC90 and ENV is shown in figure 8.1 as a function of the
concrete strength. The figure covers indented wires and strands, different bond conditions and the
way of release. The lower design value of the transmission length is shown.
The straight line representing the ENV is the same in all 6 diagrams, since the ENV does not make
any difference between strands and indented wires, nor does it take into account bond conditions, nor
the way of release (at least does not tell how to do it). The EN and MC90, on the other hand, do take
these parameters into account.
Due to this the comparison gives very different results. For example, for strands released gradually
and in good bond conditions, the ENV is extremely conservative compared to the EN and MC90,
whereas for indented wires released suddenly and in bad bond conditions, the ENV is by far on the
unsafe side.
8.10.2.3.1.2 Effect of initial prestress
Figure 2 shows the transmission length as a function of the initial prestress. The most notable
difference in this case is that the ENV does not take into account the magnitude of the initial prestress
at all (at least it is not explained how to do), whereas according to the EN and MC90 the transmission
length is directly proportional to the prestress.
The comment to figure 8.1 in the third paragraph of 8.10.2.3.1.1 applies also to figure 8.2.
8.10.2.3.2 Anchorage length
In figure 8.3 the total anchorage length in ULS is shown as a function of the stress to be anchored.
Like for the transmission length, the ENV is conservative for gradually released strands in good bond
conditions, and far on the unsafe side for wires in bad bond conditions, particularly if released
suddenly.
Guide to EC2 Section 8
Page 85 Table of contents
Figure 8.1.Transmission length (lower design value lpt1) as a function of the
concrete strength fck(t) at release. Initial prestress is 1000 MPa.
Thick black line = EN, thick grey line = ENV, thin black line = MC90
Figure 8.2.Transmission length (lower design value lpt1) as a function of the initial prestress
ıpi. Concrete strength at release is 40 MPa. Thick black line = EN,
thick grey line = ENV,thin black line = MC90
Guide to EC2 Section 8
Page 86 Table of contents
Figure 8.3. Anchorage length lbpd as a function of the stress to be anchored, ıpd.
Concrete strength at release is 40 MPa, nominal concrete strength is C50 and
initial prestress is 1000 MPa.
Thick black line = EN, thick grey line = ENV, thin black line = MC90
Guide to EC2 Section 11
Page 111 Table of contents
SECTION 11 LIGHTWEIGHT
CONCRETE
11.1 General
11.3 Materials
SECTION 11 LIGHTWEIGHT CONCRETE
C11.1 General
A favourable circumstance for the preparation of Chapter 11 was that during the last years
substantial work was conducted in updating the stateontheart on lightweight aggregate concrete
to the most actual level. In this respect the joint CEB/FIP (now fib) Task Group 8.1 published in
1999 fib Bulletin 4 “Lightweight Aggregate Concrete: Codes and Standards”, a StateoftheArt
Report giving a good overview of common practice with regard to design practice in various
countries [1]. Another report “Structural Lightweight Aggregate Concrete: Recommended
Extensions to Model Code 90”, by the same Task Group, is now ready to be published [2]. Other
valuable work was done within the scope of the BriteEuram Project “EuroLightcon”. Important
reports produced by the partners in this project were [3] and [4].
Chapter 10 in the new draft for EC2 of 1/1/2000 applies to all concretes with a closed structure
made with natural or artificial mineral lightweight aggregates, unless reliable experience indicates
that provisions different from those given can be adopted safely.
“Lightweight aggregate concrete” is defined as a concrete having a closed structure and an oven
dry density of not more than 2200 kg/m
3
consisting of or containing a proportion of artificial or
natural lightweight aggregates having a particle density of less than 2000 kg/m
3
.
C11.3 Material properties
The material properties of lightweight aggregate concrete are related to the corresponding
properties of normal aggregate concrete as defined in section 3.1. The following conversion factors
have been introduced in order to derive the properties of lightweight concrete from those of normal
weight concrete:
qE conversion factor for the calculation of the modulus of elasticity
q1 coefficient for the determination of the tensile strength
q2 coefficient for the determination of the creep coefficient
q3 coefficient for the determination of the drying shrinkage
µ ovendry density of lightweight aggregate concrete in kg/m
3
Concrete strength and stressstrain relations
The strength classes for lightweight aggregate concrete range from LC 12/15 to LC 80/95, where
the first figure stands for the characteristic cylinder strength and the second for the characteristic
cube strength.
Altogether 3 stress strain relations have been defined. The first one describes the average
behaviour as realistically as possible and is meant for calculating the distribution of forces and
moments in a structure. The second and the third are both given, as alternatives, for the design of
cross sections. The second and the third differ only in the ascending branch, the second being
parabolic and the third being linear, Fig. 11.1.
Lightweight concrete is more brittle than normal density concrete of the same strength class. This is
reflected in the formulation of the ultimate strain, where the factor q1 has been added.
Figure 11.1. Design stressstrain relations for concrete in compression
The maximum stress in the diagram is obtained by multiplying the characteristic cylinder strength
with a sustained loading factor of 0.85 and dividing by a material safety factor of 1.5, so that
fcd = 0.57fck. For normal density concrete the sustained loading factor is defined to be 1, unless
Guide to EC2 Section 11
Page 112 Table of contents
specified otherwise (Clause 3.5(1P)). The most important reason is that sustained loading effects, if
occurring anyhow, will occur after a considerable time. So when loaded the concrete will be much
older (maybe even years) than 28 days. The sustained loading effect is therefore with high
probability compensated by the gain in strength between 28 days and actual loading of the
structural member (see further the background report of chapter 3.1).
In lightweight concrete, however, the increase in strength after 28 days is smaller than in normal
weight concrete [4]. Furthermore it is reported that the sustained loading effect is more pronounced
than in normal aggregate concrete. Weigler [5] reported that the strength of lightweight concrete
under sustained loading was only about 7075% of the short term strength. Similar results were
obtained by Smeplass [6]. The results are explained by creep of the matrix, overloading the
aggregates. Consequently this phenomenon occurs when the strength of the aggregates it utilized
to its maximum [4]. Since further research seems to be necessary here, this would support the idea
of introducing a sustained loading factor for lightweight concrete anyhow. Therefore in 10.3.1.5 olcc
should preferably be defined as 0.85 “unless specified otherwise”.
The tensile strength of lightweight concrete can be obtained by multiplying the corresponding
strength of normal density concrete of the same strength class with a factor
q1 = 0.40 + 0.60µ/2200 (11.1)
where µ is the upper limit of the ovendry density.
Here it should be noted that the average tensile strength fctm normal density concrete follows from
fctm = 0.30 fctk
2/3
for concretes <C50/60 (11.2a)
and
fctm = 2.12 ln (1+fcm/10) for concretes > C50/60 (11.2b)
the characteristic (5%) value follows from
fctk = 0.7 fctm (11.3)
Emodulus
An estimate of the mean value of the secant modulus Elcm for LWAC can be obtained by multiplying
the corresponding value for normal density concrete by the coefficient
qE = (µ/2200)
2
(11.4)
In the previous part of the Eurocode, ENV 199214 “General rules for lightweight concrete with a
closed structure, the equation qE = (µ/2200)
2
was mentioned. For normal density concrete,
according to chapter 3.1, the Emodulus is calculated from
Ecm = 9.5(fck+8)
1/3
(11.5)
Creep
In the the old version ENV199214 it is denoted that for lightweight concrete the creep coefficient
 can be assumed equal to the value of normal density concrete multiplied by a factor (µ/2300)
2
for
µ > 1800 kg/m
3
. For µ < 1500 kg/m
3
a factor 1.3(µ/2300)
2
can be used. For intermediate values of
µ linear interpolation may be applied. Furthermore the creep strain has to be multiplied by a factor
q2 = 1.3 for lightweight concrete classes lower than LC20/25.
There is however serious doubt on the correctness of the statement in ENV 199214 that the creep
of lightweight concrete is smaller than that of normal density concrete, in spite of the fact that,
according to [1] also other codes like the Norwegian Code NS 3473, the Japanese Code JSCE and
the German code DIN4219 give formulations with the same tendency.
Kordina [5] states that creep is a matter of the cement paste and not of the aggregate, which would
imply that similar compositions of LWAC and NDC should give the same specific creep. Neville [6]
developed a twophase model where he distinghuishes the cement paste and the aggregates as
two parallel load bearing components. The stiffer the aggregate, the more load will be carried by the
aggregate skeleton and the more the stresses in the paste will decrease. A decrease of the
stresses in the paste will result in smaller creep deformation of the paste and hence of the concrete.
Since most of the lightweight aggregates have a lower stiffness, the stresses in the paste will
remain higher and so the creep of LWAC [4]. Anyhow, existing information seems to confirm that
there is not difference between normal density concrete and lightweight concrete with regard to the
Guide to EC2 Section 11
Page 113 Table of contents
specified creep, see f.i. Fig. 11.2 [7].
A reconsideration of the formulation for creep of LWAC, as given in ENV 199214 and provisionally
adopted in the version of EC2 of 1/1/2000 seems to be necessary. The best formulation seems to
be that creep of LWAC is the same as creep of NDC and can be calculated with the same
formula’s.
0
5
10
15
20
specific creep [10 N/mm ]
5 2
10 20 30 40 50 60 70
strength at time of loading [N/mm ]
2
NDC
LWAC
Figure 11.2. Final specific creep as a function of strength at the age of loading [7]
Shrinkage
For normal strength concrete the shrinkage is formulated as
ccs = ccd + cca (11.6)
where ccs final shrinkage strain
ccd drying shrinkage strain
cca autogenous shrinkage
The component ccd, representing the drying shrinkage strain, is known to be higher for lightweight
concrete. In [8] it is reported that the final shrinkage of LWAC is about 11.5 times the final
shrinkage of NDC of the same strength. Hoffman and Stöckl [9] reported for LWAC’s with cylinder
strengths of 4050 MPa differences of about 30% with NDC. Theissing [10] reported, on the basis
of a literature survey, for concretes with a cylinder strength of 21 Mpa, values which were about
35% higher than for NDC. Probst [11] reported about shrinkage tests on three different LWAC’s
made with Liapor (fcc = 65 Mpa), Berwilit (fcc = 43 Mpa) and Leca (fcc = 33 Mpa), kept under a RH of
65%, values of 0.55% in axial direction and 0.85 % in transverse direction. This is about the same
as found for NDC.
In ENV199214 it is stated that final drying shrinkage values for lightweight concrete can be
obtained by multiplying the values for normal density concrete with a factor q3 defined by
LC12/15 to LC/20: q3 = 1.5
LC20/25 and higher q3 = 1.2
Since the information from literature is not fully consistent and the values given in ENV 199214 will
presumably be not be too far from reality they have been maintained in the new draft of 1.1.2000.
A new element in the formulation of shrinkage is the component cca which represents autogenous
shrinkage. The attention to this additional type of shrinkage contribution was drawn during the
introduction of high strength normal density concretes, with low water/cement ratio’s. Autogenous
shrinkage is believed to be caused by “selfdissication”, which is a result of a volume reduction of
the hydration product compared to the volume of the reacting water and cement, i.e. chemical
shrinkage, and goes along with a decrease of the relative humidity in the pore system. This drop in
relative humidity is accompanied by a volume reduction of the matrix. This volume reduction is
sometimes also denoted with the term “chemical shrinkage”. [4].
For normal density concrete the component of autogenous shrinkage is formulated as:
cca,·
(t) = cc(t) cca,·
(11.7a)
where
Guide to EC2 Section 11
Page 114 Table of contents
cca,·
= 2.5 (fck – 10) 10
3
(11.7b)
and cc(t) is the hardening function.
The contribution of autogenous shrinkage decreases considerably with increasing strength of the
concrete.
In lightweight aggregate concrete the conditions are quite different if the aggregate particles are
saturated with water. In that case the possible supply of water from the aggregate to the drying
microstructure will prevent a significant drop of the relative humidity in the paste and will thus
reduce autogenous shrinkage. Therefore for LWAC the contribution of autogenous shrinkage as
given by Eq. (EC3.10/11) has to be regarded as an upper value.
Ultimate bearing capacity of LWAC structures
With regard to the bearing capacity of structures in the ultimate limit state specially the behaviour in
shear and punching is important. This holds particularly true because cracks in lightweight
concretes are supposed to be smoother than cracks in normal density concrete: in normal strength
concretes of moderate strength cracks are propagating around the aggregate particles, whereas in
lightweight concrete the crack intersects the aggregate particles, which have generally a much
lower strength than gravel aggregate particles. This might reduce the shear friction capacity of the
cracks and as such reduce the total shear carrying capacity. This difference might also limit the
redistribution capacity of the concrete web (rotation of compression struts to lower angles), which is
particularly important since in the draft of 1.1.2000 the standard method has not been involved
anymore and only the variable inclination method is given as the basis for the calculation of the
shear reinforcement. Those questions will be systematically treated.
Shear capacity of reinforced concrete members without shear reinforcement
In Section 6, the shear capacity of reinforced (normal density) concrete members without shear
reinforcement has been formulated as:
VlRdc = [0.12q1k(100µflck)
1/3
– 0.15ocd] bwd (11.8)
where
flck characteristic cylinder strength of lightweight concrete
k size factor = 1 + (200/d)
1/3
2.0
µl longitudinal reinforcement ratio = Asl /bwd < 0.02
ocd average longitudinal prestress in the cross section
q1 conversion term from NDC to LWAC, see Eq. 11.1.
The coëfficient 0.12 has been replaced (0.18/¸c) in order to show explicitly the safety margin.
On order to seen if this formulation is also suitable for lightweight concrete, the expression has been
verified with 86 test results, from Ivey/Buth [12], Walraven [13], Hansson [14], Taylor/Brewer [15],
Evans/Dongre [16], Torenfeld/Drangsholt [17], Thorenfeld/Stemland [18] and Aster/Koch [29]. Fig.
11.3 shows the results. A mean value of vtest/kq1(µfcm)
1/3
of x = 0.162 with a standard deviation of
0.0235 is obtained, corresponding to a coefficient of variation of 0.145. A design value can be
obtained on the basis of a statistical evaluation. For such a case the classical level2 method, as
described in EC1 Basis of Design is suitable. The way how to deal with this method has been
described and illustrated by Taerwe [20].
Guide to EC2 Section 11
Page 115 Table of contents
0 10 20 30 40 50 60 70 80
f
cm
0
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
0.18
0.20
0.22
0.24
0.26
0.28
v / k (100 )
u cm
q µ
1
f
1/3
Ivey, Buth [12]
Walraven [13]
Hanson [14]
Taylor, Brewer [15]
Evans, Dongre [16]
Thorenfeld, Drangshold [17]
Thorenfeld, Stemland [18]
Figure 11.3. Verification of Eq. 11.8 for the shear capacity of members
without shear reinforcement with test results
According to the level 2 method, a reliable design equation can be derived from test results with the
general formulation
BRd = µBR(1  oBr  oBR) (11.9)
where
BRd design value
µBR mean value of tests
oBR sensitivity factor for BR normally taken 0.8 in the case of one dominating parameter
 target safety index, taken 3.8
oBR coefficient of variation
with µBR = 0.162, oBR = 0.8 and oBR = 0.145 a value for the design coefficient in Eq. 11.8 of 0.091
is obtained. In this derivation, however, the mean concrete cylinder compressive strength has been
used, whereas in the code expression the 5%lower value fck is used. In the new version of EC2,
according to the Model Code, the relation
fck = fcm – 8 (Mpa) (11.10)
is used. This means coefficients of variation of = 0.15 for a concrete LC 25/30 and of = 0.055 for a
concrete LC 80/95. This would mean an increase of the coefficient 0.091 with 9% for C25 and 3%
for LC 80/95.This would then result in a coefficient 0.100 for a concrete class LC25/30 and 0.095 for
a concrete class of LC80/95. The conclusion is that Eq. 11.8 should be modified to
V1Rdc = [0.10 q1 k (100µ fck)
1/3
– 0.15ocd] bwd (11.11)
This agrees with the proposal given in [2].
Shear capacity of members with shear reinforcement
In the new version for EC2, contrary to ENV 199211, only one method for the design of members
with shear reinforcement is given. This method is based on the variable angle truss model.
For members of normal density concrete not subjected to axial forces, with vertical shear
reinforcement, the shear capacity is the smaller value of
,
cot
sw
Rd sy ywd
A
V zf
s
= u (11.12a)
where Asw = crosssectional area of one stirrup, s = stirrup distance, z = inner lever arm of the
crosssection, fywd = design yield stress of shear reinforcement, u = inclination of compression strut,
and
VRd,max = bw z v fcd / (cotu + tanu ) (11.12b)
Guide to EC2 Section 11
Page 116 Table of contents
With the additional condition
cd
w
ywd sw
vf 5 . 0
s b
f A
s (11.12c)
The first eq. 11.12a represents yielding of the shear reinforcement and the second equation 11.12b
crushing of the inclined concrete struts. The inclination of the concrete struts can freely be chosen
between 21.8
0
(cot u = 2.5) and 45
0
(cot u = 1).
v is an efficiency factor for the concrete crushing strength depending on the concrete strength
according to:
v = 0.6(1 – fck/250) 0.5 (11.13)
An important question with regard to the applicability of those formulations for lightweight aggregate
concrete is if the concrete struts in the web have a sufficient capacity to rotate. In normal density
concrete during crack formation the strong aggregate particles do not fracture and the crack
propagates around them: therefore the crack surface is very rough so that large frictional forces can
be transmitted. This is a very important condition to allow a rotation of the inclined struts from 45
0
down to an angle of 21.8° as a minimum. In lightweight aggregate concrete the aggregate particles
are intersected, so that a less rough crack surface is obtained. It is therefore questionable whether
the rotation capacity of the web is sufficient to allow as well a lowest strut inclination of 21.8°, or if a
higher lower limit should be defined. In order to answer this question tests have been carried out on
Ishaped beams with varying shear reinforcement, Fig. 11.4, Walraven [21]. Any series consisted of
three beams, which contained whether low, medium or high ratio’s of shear reinforcement, whereas
furthermore the beams were exactly similar. Three types of lightweight aggregates were used in the
various concrete mixes: Lytag, Liapor and Aardelite. In those concretes only the coarse aggregate
particles were of the lightweight type: the mixtures contained natural sand. The concrete volume
weights were 2050 kg/m
3
(Aardelite), 1975 kg/m
3
(Lytag) and 1780 kg/m
3
(Liapor). Those series
were compared with a reference series with beams made of normal density concrete. On the web
the state of deformation was continuously measured, so that the inclination of the principal
compression strain could be monitored. Fig. 11.4 shows two diagrams, in which the inclination of
the principal strain is represented. The left diagram shows the results for the gravel concrete
members, with low (GD30L), medium (GD30M) and high (GD30H) shear reinforcement ratio’s, the
right diagram shows the corresponding curves for lightweight concrete. The tests show that the
rotational behaviour of the inclined struts is similar for LWAC and NDC. In both cases the beams
with the lowest shear reinforcement ratio showed the highest strut rotation capacity. Obviously the
other two shear reinforcement ratio’s were both too high to reach yielding of the steel, so that the
final rotation remained relatively small. The unexpected result that NDC and LWACbeams behave
similarly can be explained by the overall shape of the cracks. On a mesolevel the roughness of the
cracks in lightweight concrete is indeed smaller, but this was compensated by the roughness on the
macrolevel, caused by the overall crack undulation. In this way also in the interface contact areas
occurred, with sufficient capacity to develop the necessary transmission of forces across the
inclined cracks. Also Thorenfeld [22] reported a substantial decrease of the strut inclination with
increasing load. For a shear reinforcement ratio of 0.5% he found a lowest strut inclination at failure
of 25
0
. His tests were carried out on lightweight concrete with Leca aggregates, both for the course
and the sand fractions. The volume weight of this concrete was 1500 kg/m
3
.
20 20
25 25
30 30
35 35
40 40
45 45
50 50
u u
0 100 200 300 400 500 600 0 100 200 300 400 500 600
V [kN] V [kN]
gravel
H
H
M
M
L
L
lightweight
Figure 11.4. Principal strain directions u in relation to the longitudinal member axis, as a function of the shear force for
gravel concrete (left) and lightweight concrete (Aardelite, right),
for high (H), medium (M) and low (L) shear reinforcement ratio’s [21]
On the basis of those observations, for the design of members with shear reinforcement in
lightweight concrete, the same principle as for normal density concrete was maintained in the new
version for EC2, including a lower limit for the strut inclination of 21.8° (cot u = 2.5).
Guide to EC2 Section 11
Page 117 Table of contents
In ENV 199214, the part on lightweight aggregate concrete, the efficiency factor v, defining the
crushing capacity of the concrete struts, was formulated as
v = 0.6 – flck/235 0.425 (11.14)
The efficiency factor v is only slightly smaller than the corresponding expression for NDC in the new
EC2 version. Comparison with tests shows that the combination of the equations 11.12.ac and
11.14 does not give an appropriate lower bound. An analysis showed that this can not be solved by
restricting the strut rotation to a higher value of umin. reducing the allowable inclined compressive
stress. Implicitly this means that a reduction of the efficiency factor v is necessary. A better
formulations is therefore
vLWAC = 0.85 q1 vNDC (11.15a)
or
vLWAC = 0.85 q1 0.6 (1 – fck/250) 0.425 q1 (11.15b)
Fig. 11.5 shows the comparison of the combination of Eq. 11.12ac and 11.15 with test results from
Walraven [21], Hamadi [22] and Thorenfeld [23].
0
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
v / vf
u cm
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
µf / vf
sw cm
Walraven [20]
Hamadi [21]
Thorenfeld [18]
Figure 11.5. Verification of the design equation 11.12/15 for lighweight concrete members
with shear reinforcement by test results
Eq.11.14 could be simplified to
vLWAC = 0.50(1 – fck/250) (11.16)
Punching shear capacity of lightweight concrete slabs and column bases.
In normal density concrete the punching shear capacity is calculated with the nominal design
punching shear stress
vRd,c = 0.12Ș1k (100 µl fck)
1/3
– 0.08 ocd (11.17)
which is multiplied with a basic control section at a distance of 2d from the loaded area.
Since Eq. 11.17 is the same as the equation used for shear, it may be wondered whether also here,
like in the case of Eq. 11.9, a factor 0.10 should be applied in stead of the factor 0.12 basically valid
for NDC. Fig. 11.6 shows a comparison between Eq. 11.17 (only reinforced concrete, so ocd = 0)
and test results by Tomaszewicz [24], Regan [25]. Hognestad [26], Corley [27] and Ivey [28].
Although the number of available tests was limited, a statistical derivation according to the level 2
method leads to the conclusion that the coefficient 0.12 is correct, so that Eq. 11.16 can be
maintained (for the 22 tests a mean value of µRd = vtest/vcalc =0.181 is obtained, with a standard
deviation of 0.0183, which means a coefficient of variation of 0.10. The level2 method yields then a
design value, including the model uncertainty, of xd = 0.181 – 0.8ā3.8ā 0.0183 = 0.125).
Guide to EC2 Section 11
Page 118 Table of contents
0 10 20 30 40 50 60 70
f [MPa]
cm
0.0
0.05
0.10
0.15
0.20
0.25
v / k (100 f )
u,exp cm
q µ
1
1/3
Tomaszewicz [24]
Regan [25]
Hognestad [26]
Corley [27]
Ivey [28]
Figure 11.6. Verification of Eq. 11.16 for the punching shear capacity of slabs
with test results modification
Multiaxial states of stress
A short review of the major points on this subject, as treated in [1] and [2], is given by Faust in [30].
The text is cited here:
Major differences between LWAC and NDC arise from differences in transversal behaviour,
strength under multiaxial state of stress, local compression and efficiency of confinement. All of
these aspects are connected with each other and attributed to various LWAC phenomena. First, the
transversal strain of LWAC at the maximum compressive load is in part considerably lower,
although Poissons ratio in the elastic region is almost the same as for NDC. This is valid in
particular for LWAC with lightweight sand, because of the minimum microcracking. The second
reson is the lesser resistance of LWA to lateral pressure in comparison to dense aggregates. This
leads to a higher compressibility under multiaxial loads due to the porous nature of LWA. Hence,
increasing effects of lateral pressure on confined concrete sections are generally reduced in LWAC,
Fig. 11.7.
Finally, the lower ratio between the tensile and the compressive strengths reduces the compressive
bearing capacity of locally loaded LWAC (Fig. 11.8).
Figure 11.7. Efficiency of a confining reinforcement in different concretes according to MC’90, fib Bulletin 8 and CEB
Bulletin 228
Guide to EC2 Section 11
Page 119 Table of contents
Fig. 11.8. Local compression in LWAC according to Walraven et al. [31]
Deflection of slabs
Since the Emodulus for LWAC is smaller than for NWAC this will influence the acceptable limits for
l/d. This can be calculated as follows:
The deflection of a slab, spanning one way, and subjected to a uniformly distributed load is
(assuming the cracked state)
r ) EI (
ql
384
5
4
= o
with Mmax = 1/8 ql
2
this can be written as:
r ) EI (
l M
48
5
2
max
= o
with
r
max
k
r ) EI (
M
= and
) x d ( E x d
k
s
s s
r
÷
o
=
÷
c
=
it is found that
)
d
x
1 ( d E
l
48
5
s
2
s
÷
o
= o (11.18)
where os is the stress in the longitudinal steel and x is the height of the compression zone.
In general the requirement į 0.004 l will be governing.
Substituting this equation in (11.1) it is found that
s
s
)
d
x
1 ( E 038 . 0
d
l
o
÷
< (11.19)
For the height x of the compression zone in the linear elastic cracked state it is known that
1
2
1 1
n 2 ) n ( n
d
x
µ + µ + µ ÷ = (11.20)
where µ1 = longitudinal reinforcement ratio and n = Es/Ec
For a concrete strength C20/25 the Emodulus, including creep effects, can be assumed to be
Ec ~ 9000 MPa, so that n ~ 200,000/9000 = 22,2
For lightweight concrete Elc= qE Ec, where qE is defined as
2
E
2200

.

\
 µ
= q , see Eq. 11.2 in prEN199211 (11.21)
In Table 11.1 the values for x/d are calculated on the basis of Eq. (11.19), as a function of µ1(%)
and qE
Guide to EC2 Section 11
Page 1110 Table of contents
Table 11.1 values for x/d calculated on the basis of Eq. (11.19), as a function of µ1(%) and qE
µ(kg/m
3
) 2200 2000 1800 1600
qE 1 0,82 0,67 0,53
µ1 = 0.2%
26 . 0
d
x
= 0.235 0.217 0.20
0.3%
30 . 0
d
x
= 0.28 0.26 0.24
0.4%
34 . 0
d
x
= 0.31 0.29 0.27
0.5%
37 . 0
d
x
= 0.34 0.32 0.29
1.0%
48 . 0
d
x
= 0.45 0.42 0.38
The values x/d being known as (1x/d) is known; on the basis of Eq. 11.19 the change of l/d can be
calculated.
Table 11.2 gives the reduction factors for l/d obtained in this way
Table 11.2. reduction factors for l/d
Reduction factor R for l/d
qE=1 qE=0.82 qE=0.67 qE=0.53
µ1 = 0.2 1 0.974 0.950 0.925
0.3 1 0.972 0.946 0.921
0.4 1 0.96 0.93 0.90
0.5 1 0.95 0.93 0.89
1.0 1 0.95 0.89 0.84
Fig. 11.9 shows the results graphically. In practical design the reinforcement ratio is generally
between 0.15 and 0.50%. It can be seen that the reduction factor
15 . 0
E
q covers the calculated
values quite well.
Figure 11.9. Reduction factor R for l/d as a function of qE and µ1
Guide to EC2 Section 11
Page 1111 Table of contents
REFERENCES
1. fibTask Group 8.1, “Lightweight Aggregate Concrete: Codes and Standards”, StateoftheArt
Report , Lausanne, Aug. 1999, 40 pages
2. fibTask Group 8.1,”Structural Lightweight Aggregate Concrete: Recommended Extensions to
Model Code 90”, 35 pages, to be published in 2000.
3. European Union – Brite Euram III, “Definitions and International Consensus Report”,
Eurolightcon Project, Document BE963942/R1, April 1998.
4. European Union – Brite Euram III, LWAC Material Properties StateoftheArt”, Document
BE963942/R2, December 1998.
5. Kordina, K, “Experiments on the influence of the mineralogical character of aggregate on the
creep of concrete”, Rilem Bulletin, Paris, 1960, No. 6, pp 722.
6. Neville, A.M., Dilger, W.H., Brooks, J.J.,”Creep of plain and structural
concrete”Construction Press, London.
7. CEBFIP Manual on Lightweight Aggregate Concrete, the Construction Press, London, 1977
8. Cembureau (1974), Lightweight aggregate concrete – Technology and World applications,
Editor G. Bologna.
9. Hoffmann, P., Stöckl, S., “Tests on creep and shrinkage of high strength lighweight aggregate
concrete”, Deutscher Ausschuss für Stahlbeton, Nr. 343, pp. 120 (in German).
10. Theissing, E.M., et al. “Lightweight concrete”, CUR report 48, 208 p., 1971.
11. Probst, P., Stöckl, S.,”Experiments on creep and shrinkage of high strength lightweight
concrete”, Deutscher Ausschuss für Stahlbeton, Nr. 313, pp. 5881 (in German).
12. Ivey, D.L., Buth, E.,”Shear capacity of lightweight concrete beams”, ACI Journal, Proceedings,
Oct. 1967.
13. Walraven, J.C.,”The influence of depth on the shear strength of lightweight concrete beams
without shear reinforcement”, Report 5784, Delft University of Technology, 1978.
14. Hanson, J.A., Tensile strength and diagonal tension resistance of structural lightweight
concrete, ACI Journal Proceedings, Vol. 58, July 1961, pp. 140.
15. Taylor, R., Brewer, R.S., “The effects of the type of aggregate on the diagonal cracking of
reinforced concrete beams”, Magazine of Concrete Research, July 1963, pp. 8792.
16. Evans, R.H., Dongre, A.V.,”The suitability of lightweight aggregate (Aglite) for structural
concrete”, Magazine of Concrete Research, Vol 15, No. 44, July 1963, pp. 9397.
17. Torenfeld, E.,Drangshold, G.,”Shear Capacity of Reinforced High Strength Concrete Beams”,
High Strength Concrete, Second International Symposium, May 1990, Berkely 1990, ACISP
121.
18. Thorenfeld, E., Stemland, H.,”Shear capacity of lightweight concrete beams without shear
reinforcement”, International Symposium on Structural Lightweight Concrete, 2024 June 1995,
Sandefjord, Norway, Proceedings, pp. 244255.
19. König, G., Fisher, “Model uncertainties concerning design equations for shear capacity of
concrete members without shear reinforcement”, CEBBulletin 224, Model Unvertainties, July
1995, pp. 100.
20. Taerwe, L.,”Towards a consistent treatment of model uncertainties in reliability formats for
concrete structures”, CEB Bulletin 219, 1993,pp. 561.
21. Walraven, J.C., AlZubi, N.,”Shear capacity of lightweight concrete beams with shear
reinforcement”, International Symposium on Structural Lightweight Concrete, 2024 June 1995,
Sandefjord, Norway, Proceedings, pp. 91104.
22. Hamadi, Y.D., Regan, P.E.,”Behaviour of normal and lightweight aggregate beams with shear
cracks”, The Structural Engineer, Vol 58B, No. 4, December 1980, pp. 7179.
23. Thorenfeld, E., Stemland, H. and Tomaszewicz, A.,”Shear Capacity of Large IBeams”,
International Symposium on Structural Lightweight Concrete, 2024 June 1995, Sandefjord,
Norway, 1995, Proceedings, pp. 733744.
24. Tomaszewicz, A., High –Strength Concrete. SP2 – Plates and Shells. Report 2.3 “Punching
Shear Capacity of Reinforced Concrete Slabs”, Report No. STF70 A93082, SINTEF Structures
and Concrete, Trondheim, 36 pp.
25. Regan, P.E., AlHussaini, A., Ramdane KE., Xue HY., (1993). “Behaviour of High Strength
Guide to EC2 Section 11
Page 1112 Table of contents
Concrete Slabs”, Concrete 2000. Proceedings of International Conference, University of
Dundee, Scotland, UK, September 79, Vol. 1, E&FN Spon, Cambridge, pp. 761773.
26. Hognestad, E., Elstner, R.C., Hanson, J.A., “Shear strength of reinforced structural lightweight
agregate concrete slabs”, ACI Journal, June 1964, PP. 643656.
27. Corley, W.G., Hawkins, N.M., Shearhead reinforcement for slabs”, ACIJournal, October 1968,
pp. 811824.
28. Ivy, C.B., Ivey, D.L., Buth, E., Shear capacity of lightweight concrete flat slabs”, ACIJournal,
June 1969, pp. 490494.
29. Aster, H., Koch, R.,”Untersuchungen an dicken Stahlbetonplatten”, Universität Stuttgart, 1974,
30. Faust, T., “Recommended extensions to Model Code 90 for lightweight aggregate concrete”,
fib news, 2000, No. 3, pp. 153156.
31. Walraven, J.C., den Uijl. J.A., Stroband, J., AlZubi, N., Gijsbers, J. and Naaktgeboren, M.,
“Structural Lightweight Concrete: Recent Research”, Heron, 1995, 40, No. 1, pp. 5 – 30.
See example 11.1, 11.2
EUROCODE 2 COMMENTARY
photocopying. All advice or information from the European Concrete Platform ASBL is intended for those who will evaluate the significance and limitations of its contents and take responsibility for its use and application. stored in a retrieval system or transmitted in any form or by any means. recording or otherwise. Whitman). prof J. the European Concrete Platform ASBL cannot guarantee either. during the preparation of the EN version of Eurocode 2 (prof A. and on background documents prepared by the Eurocode 2 Project Teams Members. Readers should note that all European Concrete Platform publications are subject to revision from time to time and therefore ensure that they are in possession of the latest version. prof R. Westerberg. without the prior written permission of the European Concrete Platform ASBL. No liability (including for negligence) for any loss resulting from such advice or information is accepted. Walraven. June 2008 All rights reserved. on behalf of the the Italian Cement Organziation AITEC.V. electronic. the Italian Association for Reinforced and Prestressed Concrete. . The opinions reflected in this document are those of the authors and the European Concrete Platform ASBL cannot be held liable for any view expressed therein. Corres Peiretti. No part of this publication may be reproduced. This publication is based on the publication: "Guida all'uso dell'eurocodice 2" prepared by AICAP. Beeby. While the goal is to keep this information timely and accurate. prof B. Information on European Concrete Platform document does not create any liability for its Members. It is given in good faith. they will be corrected. Authorization has been received or is pending from organisations or individuals for their specific contributions.COMMENTARY EUROCODE 2 Copyright: European Concrete Platform ASBL. mechanical.W. Published by the European Concrete Platform ASBL Editor: JeanPierre Jacobs 8 rue Volta 1050 Brussels Layout & Printing by The European Concrete Platform All information in this document is deemed to be accurate by the European Concrete Platform ASBL at the time of going into press. prof H. If errors are brought to its attention.
.
.
Eurocodes reflect the results of research in material technology and structural behaviour in the last fifty years and they incorporate all modern trends in structural design. which offers tools for the design of economic and innovative concrete structures. Also EC 2 in common with other Eurocodes. The chapter on EN 1990 (Basis of structural design) is an added bonus and will be appreciated by practioners. EFCA and ERMCO this publication should prove immensely valuable to designers in discovering the background to many of the code requirements. Commissioned by CEMBUREAU. BIBM. The problems of coming to terms with a new set of codes by busy practising engineers cannot be underestimated. The commentary will prove an authentic companion to EC 2 and deserves every success. The publication brings together many of the documents produced by the Project Team during the development of the code. Professor R S Narayanan Chairman CEN/TC 250/SC2 (2002 – 2005) . This publication will assist in building confidence in the new code. Worked examples further illustrate the application of the code and should promote understanding. Like many current national codes in Europe. tends to be general in character and this might present difficulty to some designers at least initially. With a wealth of code writing experience in Europe.Attributable Foreword to the Commentary and Worked Examples to EC2 Eurocodes are one of the most advanced suite of structural codes in the world. This is the backdrop to the publication of ‘Commentary and Worked Examples to EC 2’ by Professor Mancini and his colleagues. They are born out of an ambitious programme initiated by the European Union. The document is rich in theoretical explanations and draws on much recent research. it was possible to approach the task in a rational and logical manner. Eurocode 2 (EC 2) for concrete structures draws heavily on the CEB Model Code. Comparisons with the ENV stage of EC2 are also provided in a number of cases. might obscure the similarities to many national codes. And yet the presentation and terminology. They embody the collective experience and knowledge of whole of Europe. conditioned by the agreed format for Eurocodes.
A code should be simple enough to be handled by practicing engineers without considerable problems. Bo Westerberg. Here the word “accuracy” should be well understood. but not at the cost of too complex theoretical formulations.Foreword to Commentary to Eurocode 2 and Worked Examples When a new code is made. a number of principles should be regarded: 1. but not on the cost of significant concessions with regard to quality. excluding others. derived by scientists. cannot lead to very accurate results. New developments should be recognized as much as possible. corresponding to a good representation of the structural behaviour and of the material physics. or an existing code is updated. Codes should be transparent. As an alternative more detailed design rules may be offered. On the other hand simplicity should not lead to significant lack of accuracy. another important condition applies. It is a rule for every project. It is important that this background information is well documented and practically available. because the input values can not be estimated with accuracy. 5. practical rules can be given. consuming more calculation time. especially Robin Whittle. 2. consistent and coherent. that the code is not prepared for those who make it. That means that the writers should be aware. This book may. Models with different degrees of complexity may be offered. for helping in getting together all background information. International consensus had to be reached. Also my colleague Giuseppe Mancini and his Italian team are gratefully acknowledged for providing a set of very illustrative and practical working examples. Often socalled “accurate” formulations. but for those who will use it. Finally I would like to thank CEMBURAU. 3. It contains extensive background information on the recommendations and rules found in EC2. BIBM. For instance simple. 4. Codes should be based on clear and scientifically well founded theories. like EC2. I would like to thank my colleagues of the Project Team. Hugo Corres and Konrad Zilch. A code may have different levels of sophistication. A lot of effort was invested to achieve all those goals. For writing a Eurocode. that it should not be considered as finalized if implementation has not been taken care of. which means that it cannot be based on one certain theory. as such increasing the transparency. EFCA and ERMCO for their initiative. but resulting in more accurate and economic results. Joost Walraven Convenor of Project Team for EC2 (1998 2002) . 6. further to courses and trainings on a national and international level. leading to conservative and robust designs. A code should be openminded. serve as an essential and valuable contribution to this implementation. support and advice to bring out this publication.
.
8.3.1 .3 of EN 199211 without “C”: first paragraph of the comment on clause 5.1 of EN 199211 Comment on Clause 5.Eurocode 2 Commentary Reading key Clause 2.1 of the EN 199211 Comment on Clause 2.8.
1 BENDING WITH OR WITHOUT AXIAL FORCE .51 5.4 LINEAR ELASTIC ANALYSIS .2 SHEAR.2 REINFORCING STEEL .534 5.Guide to EC2 Summary COMMENTARY TO EUROCODE 2 .7 NONLINEAR ANALYSIS .210 2.312 3.1 GENERAL .4 PRESTRESSING DEVICES .52 5.6 SUPPLEMENTARY REQUIREMENTS FOR FOUNDATIONS .534 5.61 6.3 IDEALISATION OF THE STRUCTURE.7 REQUIREMENTS FOR FASTENINGS .214 SECTION 3 MATERIALS .55 5.8 ANALYSIS OF SECOND ORDER EFFECTS WITH AXIAL LOAD .21 2.26 2.51 5.54 5.1 CONCRETE .31 3.25 2.2 PRINCIPLES OF LIMIT STATE DESIGN .10 PRESTRESSED MEMBERS AND STRUCTURES .6 PLASTIC ANALYSIS .214 2.9 LATERAL INSTABILITY OF SLENDER BEAMS .311 3.4 VERIFICATION BY THE PARTIAL FACTOR METHOD .5 DESIGN ASSISTED BY TESTING .31 3.56 5.21 2.41 SECTION 5 STRUCTURAL ANALYSIS .313 SECTION 4 DURABILITY AND COVER TO REINFORCEMENT .11 ANALYSIS FOR SOME PARTICULAR STRUCTURAL MEMBERS .2 GEOMETRIC IMPERFECTIONS .1 REQUIREMENTS .214 2.612 Page 11 Table of contents .SUMMARY SECTION 1 SYMBOLS .3 BASIC VARIABLES .61 6.55 5.54 5.3 PRESTRESSING STEEL .534 SECTION 6 ULTIMATE LIMIT STATES (ULS) .11 SECTION 2 BASIS OF DESIGN .51 5.5 LINEAR ANALYSIS WITH LIMITED REDISTRIBUTION.
111 11.81 8.81 8.2 SPACING OF BARS .6 ANCHORAGES AND LAPS .8 FATIGUE .81 8.632 6.632 SECTION 7.71 7.632 6.81 8.1 GENERAL .626 6.4 DEFLECTION CONTROL .GENERAL .81 8.71 7.81 SECTION 11 LIGHTWEIGHT CONCRETE 111 11.1 GENERAL .81 8.10 PRESTRESSING TENDONS .9 BUNDLED BARS .632 6.625 6.5 ANCHORAGE OF LINKS AND SHEAR REINFORCEMENT .6 ANCHORAGE BY WELDED BARS .7 PARTIALLY LOADED AREAS .81 8.3 TORSION .81 8.81 8.81 8.71 7.71 7.4 PUNCHING . SERVICEABILITY LIMIT STATES (SLS) .3 PERMISSIBLE MANDREL DIAMETERS FOR BENT BARS .715 SECTION 8 DETAILING OF REINFORCEMENT AND PRESTRESSING TENDONS .111 Page 12 Table of contents .3 CRACK CONTROL .8 ADDITIONAL RULES FOR LARGE DIAMETER BARS .3 MATERIALS .Guide to EC2 Summary 6.1 GENERAL.5 DESIGN WITH STRUT AND TIE MODELS .4 ANCHORAGE OF LONGITUDINAL REINFORCEMENT .7 LAPS AND MECHANICAL COUPLERS .2 STRESS LIMITATION .
Note: the notation used is based on ISO 3898:1987 Latin upper case letters A Accidental action A Cross sectional area Cross sectional area of concrete Ac Area of a prestressing tendon or tendons Ap As Cross sectional area of reinforcement As.Guide to EC2 Section 1 SECTION 1 SYMBOLS SECTION 1. Ec(28) Tangent modulus of elasticity of normal weight concrete at a stress of c = 0 and at 28 days Ec. I or L beams Diameter. Depth Effective depth of a crosssection Largest nominal maximum aggregate size Eccentricity Compressive strength of concrete Table of contents Page 11 . or actual flange width in a T or L beam Width of the web on T. the following symbols apply. SYMBOLS For the purposes of this document.eff Effective modulus of elasticity of concrete Design value of modulus of elasticity of concrete Ecd Secant modulus of elasticity of concrete Ecm Ec(t) Tangent modulus of elasticity of normal weight concrete at a stress of c = 0 and at time t Design value of modulus of elasticity of prestressing steel Ep Es Design value of modulus of elasticity of reinforcing steel E Bending stiffness EQU Static equilibrium F Action Design value of an action Fd Characteristic value of an action Fk Characteristic permanent action Gk Second moment of area of concrete section L Length M Bending moment MEd Design value of the applied internal bending moment N Axial force Design value of the applied axial force (tension or compression) NEd P Prestressing force P0 Initial force at the active end of the tendon immediately after stressing Qk Characteristic variable action Characteristic fatigue load Qfat R Resistance S Internal forces and moments S First moment of area SLS Serviceability limit state T Torsional moment TEd Design value of the applied torsional moment ULS Ultimate limit state V Shear force Design value of the applied shear force VEd Latin lower case letters a a a b bw d d dg e fc Distance Geometrical data Deviation for geometrical data Overall width of a crosssection.min minimum cross sectional area of reinforcement Cross sectional area of shear reinforcement Asw D Diameter of mandrel DEd Fatigue damage factor E Effect of action Ec.
1 fp0.2k Characteristic 0. taking account only of uncertainties in the material property Increment/redistribution ratio Reduction factor/distribution coefficient Compressive strain in the concrete Compressive strain in the concrete at the peak stress fc Ultimate compressive strain in the concrete Strain of reinforcement or prestressing steel at maximum load Characteristic strain of reinforcement or prestressing steel at maximum load Angle Slenderness ratio Table of contents . having area Ac u. ratio Angle. F Partial factor for fatigue actions Partial factor for fatigue of concrete Partial factor for permanent actions.2% proofstress of reinforcement Tensile strength of reinforcement ft Characteristic tensile strength of reinforcement ftk fy Yield strength of reinforcement Design yield strength of reinforcement fyd fyk Characteristic yield strength of reinforcement Design yield of shear reinforcement fywd h Height h Overall depth of a crosssection i Radius of gyration k Coefficient. Q Partial factor for reinforcing or prestressing steel Partial factor for reinforcing or prestressing steel under fatigue loading Partial factor for actions without taking account of model uncertainties Partial factor for permanent actions without taking account of model uncertainties Partial factors for a material property. P Partial factor for variable actions.1% proofstress of prestressing steel fp0.1k Characteristic 0.Guide to EC2 Section 1 fcd Design value of concrete compressive strength Characteristic compressive cylinder strength of concrete at 28 days fck fcm Mean value of concrete cylinder compressive strength Characteristic axial tensile strength of concrete fctk Mean value of axial tensile strength of concrete fctm Tensile strength of prestressing steel fp fpk Characteristic tensile strength of prestressing steel 0. taking account of uncertainties in the material property itself. Factor l (or l or L) Length. in geometric deviation and in the design model used Partial factor for actions associated with prestressing. Span m Mass.1% proofstress of prestressing steel f0. reduced moment n reduced axial force r Radius 1/r Curvature at a particular section t Thickness t Time being considered The age of concrete at the time of loading t0 u Perimeter of concrete crosssection. coefficient Partial factor Partial factor for accidental actions A Partial factor for concrete Partial factor for actions.fat Jf Jg Jm ] c c1 cu u uk O Page 12 Angle.fat JC.z Coordinates z Lever arm of internal forces Greek lower case letters D E J JA JC JF JF.fat JG JM JP JQ JS JS. G Partial factor for a material property.y.v.w Components of the displacement of a point x Neutral axis depth x. ratio.
defining creep between times t and t0.Guide to EC2 Section 1 Coefficient of friction between the tendons and their ducts Poisson's ratio Strength reduction factor for concrete cracked in shear Ratio of bond strength of prestressing and reinforcing steel Ovendry density of concrete in kg/m3 Value of relaxation loss (in %).t0) Creep coefficient.t0) Final value of creep coefficient \ Factors defining representative values of variable actions for combination values \0 for frequent values \1 for quasipermanent values \2 P Q Q [ Page 13 Table of contents . at 1000 hours after tensioning and at a mean 1000 temperature of 20°C Reinforcement ratio for longitudinal reinforcement l Reinforcement ratio for shear reinforcement w Compressive stress in the concrete c Compressive stress in the concrete from axial load or prestressing cp Compressive stress in the concrete at the ultimate compressive strain cu cu W Torsional shear stress I Diameter of a reinforcing bar or of a prestressing duct In Equivalent diameter of a bundle of reinforcing bars (t. related to elastic deformation at 28 days (.
means the capacity of a structure to respond to a given stress. strengths and geometrical dimensions of the structure.2 Reliability management EC2 points 2.3 refer to EN1990 section 2 for all rules in relation to reliability. By the definition given by EN1990. below a fixed value. Every situation which is dangerous for a construction is referred to as a “limit state”. and finally if resistance. Finally. a crack width limit state with limited width is a reversible limit state. i. if the crack width is high. once the actions are removed the cracks cannot close). EN 1992 has some additional requirements. it includes structural safety. Exceeding the first causes collapse of the whole structure or of part of it. will be referred to later.S d 0` = P ^MS d 0` . whereas one defined by a high width is irreversible (in fact.2 and 2.2 Reliability management C2. cracking. in the second. case. exceeding the second causes limited damage that makes the structure unfit for the requirements of the project. on the other hand. geometry. no consequences of actions exceeding the specified service requirements will remain once those actions are removed.1. Once a construction has attained this condition. the basic requirements of EN1990 Section 2 are deemed to be satisfied for all concrete structures if limit state design is carried out with the partial factor method in accordance with EN1990. We recall that by ‘stress’ we mean any effect produced in the structural members by actions applied or by any other effect such as strain.e. These rules. or MS = R . or alternatively by a safety margin MS. defined as the ratio between strength R and stress S. This value is determined as a function of type of construction. Pf represents the probability that failure arises. Limit states are of two types: ultimate limit states and serviceability limit states depending on the gravity of their consequences. Given the random nature of quantities involved in structural design (actions. restraints. as well as the basic concepts of structural reliability. let us define S and R as two random variables representing respectively stress and strength.1 Requirements Eurocode 2. defined as the difference between R and S: either FS = R/S .).1. probability that a certain danger condition is attained or exceeded. it is no longer able to fulfil the functions for which it has been designed.1. and if actions are defined in accordance with EN1991. 2. structural reliability is the ability of a structure or a structural member to fulfil the specified requirements for which it has been designed. that the considered limit state is attained or exceeded at least once during T. Exceeding serviceability limit states can be reversible or irreversible: in the first case. design working life and quality management measures. increase of reinforcing steel corrosion.1 states that concrete structures should be designed in accordance with the general rules of EN1990 and with actions defined in EN 1991. durability and serviceability are dealt with in accordance with EN1992. Part 1. the probability of failure is related to a fixed reference period of time T through one of the following expressions: Pf = P ^R / S d 1` = P ^FS d 1` or Pf = P ^R .1 Requirements SECTION 2 BASIS OF DESIGN C2. the assessment of structural reliability cannot be set up by deterministic methods. For example.1. but requires a probabilistic analysis. A rigorous assessment of structural safety against a relevant limit state can be carried out by first introducing a safety factor FS. Both these factors are random variables like R and S. ‘Strength’.Guide to EC2 Section 2 SECTION 2 BASIS OF DESIGN 2. Because of the difficulty of calculation and of the limitation of available data (data which often fail to give the probabilistic distributions Page 21 Table of contents .S . The distribution of Fs or Ms is then determined on the basis of the statistical distribution of actions. some consequences will remain. This analysis (known as level 3 method) is very complex. Section 2. influence on safety of people and damage to goods.e. also taking account of the randomness of the structural scheme. serviceability and durability. i. and if combinations of actions in accordance with EN1990. For a given limit state. In particular. etc. The objective of safety verification is therefore to keep failure probability. strength of materials.
this method is of limited applicability to the design practice.Guide to EC2 Section 2 necessary for calculation). but not their statistical distributions. In circumstances where R and S are not correlated (note that in case of normal distributions noncorrelation is equivalent to statistical independence). is expressed as follows: = RS . the probability of failure can be estimated based on a index. called the “reliability index”. Alternatively. Assuming that MS is linear. are known. if only the first and second order moments (averages and standard deviations) of the random variables R and S. it was first defined by Cornell as the ratio between the average value M of MS and its standard deviation M: = M / M .
is given by the distribution function MM of MS calculated in 0: Pf = P ^MS d 0` = ) M 0 . S are the averages and standard deviations of R and S. This method (known as “level 2” method or “ method”) does not generally allow assessment of the probability of failure./ 2 R + 2 S . R . S.e. with the exception of the particular case where the relation between MS and the random variables of the problem is linear and the variables have normal distribution. where R . The probability of failure. i. the probability that the safety margin MS assumes nonpositive values.
Introducing m as the normalized variable of the safety margin MS,
m= MS M M
the result: MS =
M
+ m
M
substituted in the expression of Pf gives:
M
Pf = P ^MS d 0` = P ^ M + m M d 0` = P ^m d  M / where m indicates the distribution function of m.
` = P ^m d  ` = ) m (
) = 1 ) m ( )
The reliability index may be expressed in geometrical terms. In fact, if we introduce the normalized strength and stress variables [ r = R  R
/ R and s = S  S
/ S ], the limit condition (MS = 0) is represented in the r  s plane by a line that divides the plane into a safe region and an unsafe region (Figure 2.1). The distance from the origin of the axis of this line equals the reliability index (in circumstances where R and S are not correlated), so the verification of safety is carried out by assigning a given value to this distance.
Figure 2.1. Geometric representation of the reliability index in the rs plane
The level 2 method is also difficult to apply in practical design because the necessary data are often not available, so that another method is used: the partial factor method or semiprobabilistic method (level 1 method). This method is based on the compliance with a set of rules that ensure the required reliability of the structure by using “characteristic values” of the problem variables and a series of “safety elements”. These are represented by partial safety factors, J which cover the uncertainties in actions and materials, and by additional elements for uncertainties in geometry, e.g. to allow for the randomness of cover to reinforcement and therefore of the effective depth of a reinforced concrete section.
Page 22 Table of contents
2). e. Table 2. Table 2. protection against corrosion. 4.1. or as target values. and other civil engineering structures Structures or parts of structures that can be dismantled with a view to being reused should not be considered as temporary. in the choice of characteristic values. for example in the case of accidental actions such as impacts from road vehicles.3 Design working life. characteristic values of strength and stress are fixed as fractiles of given order of the respective distributions. measures relating to design calculations (e. The method is based on the following assumptions: 1. durability and quality management C2. including : the possible cause and /or mode of reaching a limit state.g. durability and quality management Independently from the method used for safety evaluation.g. because the probabilistic aspects of the question of safety are already taken into account in the method calibration process. the assessment of safety is positive if the design action effects don't exceed the design strengths.. measures aimed to reduce errors in design (project supervision) and execution of the structure (inspection in phase of execution) and other kinds of measures.1 gives the indicative values of design working life for different types of structures. bridges. strength and stress are independent random variables. a structure can be defined as reliable if positive safety measures have been provided for all its limit states during the whole design working life Tu. gantry girders. fixed in the Standards.EN1990] Design working life category 1 2 3 4 5 (*) Indicative design working life (years) 10 10 to 25 15 to 30 50 100 Examples Temporary structures (*) Replaceable structural parts. It has to be pointed out that the characteristic values of actions are fixed as those values with a given probability of being exceeded during the service life of the structure only if statistical data are available.1.e. i. injury. The levels of reliability relating to structural resistance and serviceability can be achieved by suitable combinations of protective measures (e. The reliability required for structures [within the scope and field of application of EN1990] shall be achieved through design in accordance with EN1990 to EN1999 and by appropriate execution and quality management measures. measures relating to quality management. bearings Agricultural and similar structures Building structures and other common structures Monumental building structures. public aversion to failure.3 Design working life. potential economical losses. the expense and procedures necessary to reduce the risk of failure. 2. characteristic values are fixed as the nominal values prescribed in standards or specifications. The choice of the levels of reliability for a particular structure should take account of the relevant factors. etc. on the basis of a given probability. choice of partial factors). Tu is defined as the period for which a structure is assumed to be usable for its intended purpose with anticipated maintenance but without major repair being necessary.1) . partial safety factors etc. 2. 3. explosions etc.1. Point (B3. EN1990 allows for the choice of different levels of reliability. both for structural resistance and for serviceability. the possible consequences of failure in terms of risk to life. Otherwise.1) of EN1990 Annex B defines three classes based on the consequences of failure or malfunction of the structure (Table 2.g. Page 23 Table of contents . protection against fire.Guide to EC2 Section 2 This method does not require that the designer has any probabilistic knowledge. Indicative design working life [Table (2. other uncertainties are taken into account by transforming characteristic values into design values. by applying partial safety factors and additional elements.).
75 107 5.g. Values for KF1 are given in Table 2. II. public buildings where consequences of failure are medium (e. with a rough probability of attaining the ULS in 50 years of 7.1. Table 4. or for economic.5 to 3.EN1990] Pf E 101 1. Recommended minimum values of the reliability index (ultimate limit states) [Table (B2)EN1990] Reliability Class RC3 RC2 RC1 minimum values at ULS 1 year reference period 50 years reference period 5.7 3.5 Design supervision and execution inspection measures being equal.2.4. Table 2. Relation between and Pf [Table C1 .3 For the serviceability limit states (irreversible). Table 2. Table 2. Table 2.g. and applying multiplication factor KF1.20 Recommended minimum values for at ultimate limit states are given in Table 2. reliability differentiation can be achieved by distinguishing classes of partial factors JF of actions to be used in combination of actions.3 gives values of for different values of Pf (remembering that the index allows the estimation of Pf values only if the relationship between MS and the random variables of the problem is linear and the variables have normal distribution). a concert hall) Residential and office buildings. for reference periods of 1 year and 50 years.9 1. storage buildings). whereas CC3 corresponds to classes III and IV.e.3).2 4. an office building) Agricultural buildings where people do not normally enter (e. social or environmental consequences considerable Low consequence for loss of human life.27 106 4. greenhouses CC1 Classes CC1 and CC2 correspond to importance classes I. public buildings where consequences of failure are high (e.7 3.5.Guide to EC2 Section 2 Table 2. Page 24 Table of contents .g. which are less dangerous and do not concern the safety of people. this can be deduced from the relation between and Pf (Table 2.5. economic.6.28 102 2. depending on the reliability index defined above.8 4. social or environmental concerns Moderate consequence for loss of human life. Target values of the reliability index for Class RC2 structural members [Table (C2)EN1990] Limit state Ultimate Fatigue Serviceability (irreversible) Target reliability index 1 year reference period 50 years reference period 4.3).2 3.3. RC3) may be associated with the three consequence classes CC1.8 for a 50 year reference period (Table 2. as from EN1998. i. Three reliability classes (RC1. different for each reliability class. CC2 and CC3.32 103 3. A design using EN 1990 with the partial factors given in annex A1 and EN 1991 to EN 1999 is considered generally to lead to a structure with a value greater than 3. Consequences classes [Table (B1) .8 1.3.2·105 (Table 2.4. economic.09 104 3. From this table it is apparent that it is the importance of the structure concerned which is the criterion for classification.72 105 4. RC2.3 4.8 2. the failure probability values for structural elements of Class RC2 are roughly 101 (1/10) in 50 years and 103 (1/1000) in 1 year.3) for the values of given in the last row of Table 2. social or environmental consequences small or negligible CC2 Examples of buildings and civil engineering works Grandstands.EN1990] Consequences Class CC3 Description Serious consequences for loss of human life.
verification of one of the two categories may be omitted only if it can be proven that it is satisfied by the verification of the other one. For the verification of ultimate limit state design actions shall not exceed the design resistance of the structure.6.1 Measures of design and execution management and quality control are aimed at eliminating failures due to gross errors. Verification shall be carried out against both categories.9 shows the ULS classification according to EN1990 [(EN1990 point (6. EC2 point 2.1. 2.8.0 1. Table 2. ULS classification Notation EQU STR GEO FAT Definition Loss of static equilibrium of the structure or any part of it considered as a rigid body. including footings. KF1 factor for actions [Table (B3)EN1990] KF1 factor for actions KF1 Reliability class RC1 RC2 RC3 0.g.2 Principles of limit state design C2..2. Design for limit states shall be based on the use of structural and load models for relevant limit states. drawings and specifications Checking performed by an organisation different from that which has prepared the design Checking by different persons than those originally Normal supervision responsible and in accordance with the procedure of the organisation. as the method for the verification of structural safety. where : minor variations in the value or the spatial distribution of actions from a single source are significant (e. Table 2.7. Page 25 Table of contents .1)]. These are here repeated. with reference to each reliability class. The verifications shall be carried out for all relevant design situations and load cases.9. partially repeating the description of the three safety verification methods in Par.7 and 2. Inspection levels (IL) [Table (B5)EN1990] Inspection levels IL3 relating toRC3 IL2 relating to RC2 IL1 relating to RC1 Characteristics Extended inspection Normal inspection Normal inspection Requirements Third party inspection Inspection in accordance with the procedures of the organisation Self inspection Design with partial factors given in EC2 and with the partial factors given in the EN1990 annexes results in a structure associated with the RC2 reliability class.Guide to EC2 Section 2 Table 2. It shall be verified that no limit state is exceeded when relevant design values for actions. Table 2. and at ensuring that the design resistance is achieved. Design supervision levels (DSL) [Table (B4)EN1990] Design supervision levels DSL3 relating to RC3 DSL2 relating to RC2 DSL1 relating to RC1 Characteristics Extended supervision Minimum recommended requirements for checking of calculations. where the strength of construction materials of the structure governs Failure or excessive deformation of the ground where the strengths of soil or rock are significant in providing resistance Fatigue failure of the structure or structural members. etc. selfweight variations) the strengths of construction materials or ground are generally not governing Internal failure or excessive deformation of the structure or structural members. or failure or excessive deformation of a structural member and they generally concern safety of people. piles. The ultimate limit states are associated with loss of equilibrium of the whole structure.9 1.2 Principles of limit state design Eurocodes adopt the partial factors method. Selfchecking: checking performed by the person who Normal supervision has prepared the design Table 2. Two categories are defined by the consequences associated with the attainment of a limit state: ultimate limit state and serviceability limit state.2 refers to EN1990 Section 3 for limit state design rules. 2. Design supervision and execution inspection levels are given at Tables 2.4. basement walls. material and product properties and geometrical data are used in these models.8. or limit states semiprobabilistic method.
3. or a set of imposed deformations or accelerations caused for example. during construction or repair. “Action” means. Each permanent action with low variability has a single characteristic value Gk. 2. either a set of forces (loads) applied to the structure (direct actions).g. waves). The definition of relevant limit states for a certain construction requires above all the analysis of the different situations to which it can be exposed. where the coefficient of variation is the ratio between standard deviation and average) and if sufficient statistical information is available. for which the variation is always in the same direction (monotonic) until the action attains a certain limit value. impact. variable actions (Q). etc.2. referring to conditions of normal use. and lower. like prestressing or concrete shrinkage. Gk. or for which the variation in magnitude with time is negligible (e.3 Basic variables 2. explosion.1 Actions and environmental influences Each design situation is characterized by the presence of several types of actions on the structure. Characteristic value of a permanent action: Gk = Gm if the coefficient of variation is negligible. The choice of combinations to be taken into account is related to the distinction between reversible and irreversible limit states: frequent and quasipermanent combinations apply to the first case. Usually the serviceability requirements are agreed for each individual project. appearance (where the term “appearance” is concerned with high deformation. Figure 2. Actions are classified as: permanent actions (G). explosions. If a permanent action has relevant uncertainties (coefficient of variation bigger than 10%. Gk.inf) should be used. which are not easily foreseeable and of low duration (e.inf is the 5% fractile of the statistical distribution for G. temperature.Guide to EC2 Section 2 Serviceability limit states correspond to conditions beyond which specified service requirements for a structure or structural member are no longer met. e. moisture variation.sup or less than Gk. by temperature changes. but also of machines and services). xaccidental situations. characteristic combinations to the second case. 2.inf is less than 5% (fig. a lower and an upper characteristic value Gk. accidental actions (A). referring to temporary conditions of the structure. where the structure is subjected to a seismic event. selfweight). This is the case of actions due to selfweight: they are generally represented through a nominal value calculated on the basis of the design drawings (structural and nonstructural member dimensions) and of the average specific gravity of materials (Gk = Gm). wind.g. fire).sup. design situations are classified as: xpersistent design situations. imposed load of people and lowduration imposed load in general on building floors). damage to finishes and to nonstructural members. and variable actions characterized by variable and nonmonotonic intensity or direction (e. Those actions. are also permanent actions.g.g. EN1990 indicates three different types of combinations for serviceability limit states verifications: characteristic combination (called “rare combination” in the previous versions of Eurocodes).2).. extensive cracking. xtransient situations. as EN1990 states. In common cases.sup are defined if the coefficient of variation is high Page 26 Table of contents . comfort of users. Gk. including fire. Exceeding these limits causes limited damage but means that the structures do not meet design requirements: functional requirements (not only of the structure. two characteristic values (upper. uneven settlement or earthquakes (indirect action). snow. the duration of which is continuous and equal to the design working life of the structure. divided in variable actions with discrete and regular occurrence in time (e.g.sup is the 95% fractile and Gk. frequent combination and quasipermanent combination. There is a 5% probability that these two values will be exceeded.inf and Gk. impacts.). which may be assumed to be Gaussian. xseismic situations. The situations chosen for design shall cover all situations that can reasonably occur during the execution and working life of the structure. the probability that the real value of action is more than Gk. etc. involving exceptional conditions of the structure or its exposure.
This period is normally coincident with the design working life of the structure. which is taken into account by its characteristic value. based on a return period of 475 years.5 Category B: office areas Category C: congregation areas Category D: shopping areas Category E: storage areas Category F: traffic area.5 0. determines the level of intensity of a variable action when this action is taken into account. as well as for service loads on building floors. the characteristic value is based upon the probability of 0.7 0. the \0 factor. DE.2 0 0 0 0. category (see EN 199111) \0 \1 Category A: domestic. In reality. \2 0. for sites located at altitude H < 1000 m a. corresponding to a probability of 10% of being exceeded in 50 years. simultaneously with another variable action. called “leading variable action”.3 resumes the representative values of variable actions. Figure 2. 2. Iceland.02 of its timevarying part being exceeded within a reference period of one year.7 0.3) and 50% (ratio of listed segments and segment AJ). Other CEN Member States. 30 kN < vehicle weight 160 kN Category H: roofs Snow loads on buildings (see EN 199113)* in Finland. HI and the reference period of 50 years represented by segment AJ in Fig. the frequent value and the quasipermanent value are inherent properties of the variable action.6 0.5 0 0. The characteristic value of a variable action has a defined probability.6 0. represented as a product \1 Qk.2 0. of being exceeded on the unfavourable side within a fixed reference period. The main representative value of a variable action is its characteristic value Qk.5 The characteristic value of seismic action for ULS verification is fixed by Eurocode 8 (EN1998). For simplicity.6 0.3 0 0. obtained by applying a reducing factor to Qk. each of these last three values is defined as a fraction of the characteristic value. Table 2.7 0.1)EN1990] Imposed loads in buildings. FG.s.2 0. vehicle weight 30 kN Category G: traffic area. in design.9 0. beforehand accepted.l.10. represented as a product \0 Qk.7 0.7 1.7 0. In other words. the frequent value.6 0.Guide to EC2 Section 2 Each variable action has four representative values. for sites located at altitude H > 1000 m a. It is possible to modify the return period by means of an importance factor JI.8 0. represented as a product \2 Qk. residential areas 0. Recommended values of \ factors for buildings [Table (A1. For the majority of climatic variable actions. Norway. It is used both for ULS verifications and for irreversible SLS verifications.s. The frequent value and the quasipermanent value of floor loads on building are determined so that the average periods of time within which they are exceeded are respectively 10% (ratios of the sum of intercepts of time BC. Page 27 Table of contents . * For countries not mentioned below. see relevant local conditions. The frequent (\1 Qk) and the quasipermanent (\2 Qk) values are used for ULS verifications including accidental actions and for reversible serviceability limit states. Wind load on buildings (see EN 199114) Temperature (nonfire) in buildings (see EN 199115) NOTE The \ values may be set by the National annex.7 0 0.10 shows the values recommended by EN1990.7 0.5 0. On the other hand. the quasipermanent value.3 0. and the \1 and \2 factors are simply the ratios between these values and the characteristic value. the other representative values are.7 0.5 0.3 0.0 0.6 0. The \0 factor takes therefore into account the low probability of simultaneous occurrence of the most unfavourable values of independent variable actions.l. Sweden and other CEN Member. Table 2. this is equivalent to a mean return period of 50 years. in decreasing order: the combination value. called the combination factor. Values of \ factors for buildings are defined in the National Annex.7 0.
due to the nature of these actions.11 shows the steps to pass from the representative values of actions to the design values of their effects on construction. Table 2. called “design values”. Schematic illustration of representative values of variable actions For accidental actions.11. but amplified values. sufficient information for the appropriate application of statistical methods is not available.3. Table 2. In order to take into account the uncertainties on the choice of characteristic values for actions and some uncertainties concerning the action modelling. Symbols representing the design values are indicated with index d. a single nominal value is determined because. design does not use characteristic values.i o \ Fk. which are obtained by multiplying characteristic values by a partial factor.i . Procedure to determine the design values of effects on structures starting from the representative values of actions Expression Fi Fk.Guide to EC2 Section 2 Figure 2.
where 0 . 2 . 1 .
2 ) by a representative values Fk. Design values of actions are determined by multiplying the 0 . quasipermanent) values.i . 1 .i .i k.i f. Fd.i (where partial factor f. Comment Actions on the structure are identified Representative values are assigned to actions: characteristic values or other (combination. frequent.i F o 0 f.i or Fk.i Fk.
2 . . where 1 f.i is a partial factor generally covering the .
i Fk. part of the uncertainties related to action modelling. 2 . values indicated on the design drawings) or data that take into account the possibility of geometrical imperfection liable to cause second order effects. action effect in a cross section).g. two different partial factors. f.i ) so that the model coefficient Sd does not explicitly appear. The design value of effects is obtained by multiplying the values produced by the design actions. the previous expression is simplified in this one.i = F.sup and G. when it is necessary to split the action into a favourable and an unfavourable part.i Fk. Actions that can occur simultaneously are considered. In normal cases. by a partial factor Sd mainly covering the uncertainties of the structural model.i or F. combinations of actions are calculated and the effects of these combinations on the structure are assessed (e. 1 . where: F. indicated as G. uncertainties related to the choice of characteristic values for actions and. sometimes. 0 . are used.i .inf . The product: Fd.i = f ( Sd . a d represents either the design value of the set of geometrical data (in general. In case of permanent actions.
a d . Ed = E f.i .i.i Fk. is often directly assumed as the design value of the action Fk.
a d . Ed = Sd E f.i Fk.i .
i Fk.a d .i . Ed = E F.
Page 28 Table of contents .
Guide to EC2 Section 1 Page 14 Table of contents .
b) multiply. Figure 2.i f( Rd . if these are not explicitly taken into account in the model.3. Following a procedure similar to the one for calculating the design value of action effects.i ¹ § X k.i ) . For the structural stiffness parameters (moduli of elasticity. The main one is strength.. Eurocode EN1990 defines the characteristic value of a property of a material as the 5% fractile of its statistical distribution where a minimum value of the property is the nominal failure limit (general case).13 gives the values of partial safety factors to be assumed for concrete and steel for ULS.1. the design value of structural resistance is determined on the basis of individual material properties and of geometrical data multiplied by a partial factor Rd that covers the model uncertainties of resistance and the geometrical data variations. thermal expansion coefficient.i ¹ Comment Material strengths and product resistances involved in the verifications are identified.). Rd 1 Rd Rd § X k.i X k. if applicable. Strength of materials is represented through a characteristic value. Procedure to determine the design values of resistances starting from the characteristic values of strength Expression Xi X k . etc.2 Material and product properties 2. The design value of a material property is determined on the basis of its characteristic value.i X d.12 shows the steps to pass from the characteristic values of individual material strengths or product resistances to the design values of structural resistance.i .i § X · R ¨ K k . This is the value that has a given probability of not being attained or exceeded during a hypothetical unlimited test series.EN1990] Page 29 Table of contents .3.1 General Several material properties are involved in structural design.4. creep coefficient.i · R¨ .a d ¸ ¨ ¸ © m. Determine the structural resistance on the basis of design values of individual material properties and of geometrical data.3.1. (C3) .3.i m. As for the action effects. i. Characteristic values of material strengths and product resistances are introduced. as well as any local defaults.e. Table 2.a d ¸ ¨ ¸ © M. factor Rd is often integrated in the global safety factor JM. according to their constituent materials. these parameters can be favourable or unfavourable. indicated as fk. and as the 95% fractile where a maximum value is the limiting value. through the two following operations: a) divide by a partial factor m .2 Thermal effects 2.i. Table 2. the characteristic value is taken as a mean value because. Relation between the single partial factors [Fig. by a conversion factor K mainly aimed at taking into account scale effects.3.4 Prestress 2.12. 2.i ¹ Fig. m. Product properties are also represented by a single characteristic value or a set of characteristic values. the ability to resist forces without breaking or failing.1.i · R¨ . to take into account unfavourable uncertainties on the characteristic of this property.2. a d ¸ ¨ J ¸ © m .3 Differential settlements/movements 2.4 summarise in a schematic way the relation between the single partial factors used in Eurocodes.Guide to EC2 Section 2 2. Table 2. by which the characteristic material strength is divided: M. in case of persistent. transient and accidental load combinations. depending on the case.
1.0 JS for prestressing steel 1.2.4.1 Partial factor for shrinkage action 2.2.1N)EC2] Design situations Persistent and transient Accidental JC for concrete 1.4.4 Partial factors for materials Table 2.3.04 VR .4 Geometric data 2.4.2.2 Supplementary requirements for cast in place piles 2.64 Vf .4.4 Verification by the partial factor method 2.4.2.4. Partial factors for concrete and steel for ultimate limit states [Table (2.Guide to EC2 Section 2 2.0 The values of partial factors given in the previous table were determined as: J M = exp 3.2 JS for reinforcing steel 1.3 Partial factor for fatigue loads 2.2 Shrinkage and creep .2.2 Partial factors for prestress 2.15 1.1 General 2.13.3.2 Design values 2. 2.4.4.15 1.3 Deformations of concrete 2.1 General 2.3.3.3.5 1.
where: VR = V 2 m 2 + VG + Vf2 .
Assuming an additional factor 1.0. the result is Jc = 1.30.50. prestressing.15 to cover the uncertainty arising from the concrete being tested with test specimens specially made and cured for this purpose. The recommended value for situations not covered by specific parts of EC2 is 1. VR Vm VG Vf coefficient of variation of resistance coefficient of variation of model uncertainty coefficient of variation of geometrical factor coefficient of variation of material strength The basic values in EC2 may be considered to be based on the following assumptions: model uncertainty Vm = 2.154 For concrete Variation of model uncertainty 5 % Variation of geometry 5 % Variation of concrete strength 15 % For these values the equation gives JM = Jc = 1. Besides the partial safety factors for materials above. Page 210 Table of contents .5 % geometry VG = 5 % steel strength Vf = 4 % For these values the equation gives JM = Js = 1. The table below gives the values recommended in EC2. These values are also defined in the National Annex. fatigue loads and materials for foundations.15·1. EC2 also defines partial factors for shrinkage. rather than from the finished structure. For reinforcement For serviceability limit states values of partial safety factors JC and JS are defined in the National Annex.30 = 1.
1).4. such as actions caused by vibrating machinery (turbines.g.3 Combinations of actions JSH JP. the variable imposed load shall be applied on the central span only. these expressions are obtained by reducing the multiplying factor of the predominant variable action (through the \0 factor) in the first case. Combinations of actions for persistent or transient design situations (fundamental combinations): ¦ jt1 G. but there is no consideration of actions of dynamic type.1 and the permanent actions (including prestressing P) introduced with its characteristic value. It has also to be noted that EN 1990 considers the zero value a possible value of the partial factor of variable actions JQ. see Example 2.i j G. the most unfavourable of which should be adopted by the designer.1 Q k. Page 211 Table of contents .g.i Q k.85 ÷ 1.1 ¦ Q.1 ¦ i !1 2. Combinations of actions for accidental design situations ¦G jt1 k.1 Q k. other concomitant variable actions introduced with their combination values.i Q k.1 )Q k.i where Ad is the accidental design action (for fire situations. j P Q.fat JC.0 1.3 1.i 0. or by reducing the multiplying factors for permanent unfavourable actions (through the [ factor that is in the range 0.i 0. A zero value of JQ is therefore an expedient to remove from the combination of actions the favourable effect of variable actions (e. for ULS verification of stability when second order effects are relevant) favourable in persistent and transient situations Stability ULS with external prestressing if an increase of Prestressing prestressing may be unfavourable local effects Fatigue Materials for foundations (amplified in order to obtain the design resistance of cast in place piles without permanent scaffolding) 2.0 1. it represents the design value of the indirect thermal action due to fire). EN1990 introduces then the possibility to choose for the fundamental combination: 1.fond 1. compressors.Guide to EC2 Section 2 Table 2. It has to be pointed out that the combinations foreseen in EN 1990 cover static actions only.unfav JP.i Q k.i or alternatively for the STR and GEO limit states.00).1 ¦ i !1 Q. 2.fav JP.i 0. which is equivalent to setting JQ = 0 in the two lateral spans.5 Partial factors for materials for foundations 2.1 0.i Q k.2. to maximize the positive bending moment in the central span of a threespan continuous beam. as defined in EN 1990 Section 6. The expression to be used for the fundamental combination is chosen in the National Annex. j G k.2 1.i where i are reduction factors for the unfavourable permanent actions G.). the more unfavourable of the two following expressions: ¦ ¦ jt1 jt1 G. or a system of two expressions. j G k.1 JC The general formats for the ULS and SLS combinations of actions. j P Ad ( 1. the traditional expression with a predominant variable action Qk.14.4. Recommended values of partial safety factors Shrinkage (e. j P P Q.1 ¦ i !1 i !1 Q. j G k. accidental and seismic. although JQ = 0 has no probabilistic meaning. etc. j P P P Q. are given below. ULS Combinations Three types of combinations of actions should be considered for Ultimate Limit States: fundamental.0 1.1 Q k.1 or 2.unfav JF.
3)EC2]: § · ¨ ¦ G k.g. j P ¦ 2. Characteristic combination ¦ G k. its frequent value (\1. traffic load or other cyclic load).i Qk. frequent and quasipermanent. limit states that will not be attained or exceeded once the actions that have caused attainment or exceeding are removed. The quasipermanent combination should be considered for long term action effects. 6).i Q k. the crack width limit state of a prestressed concrete beam with bonded tendons. the characteristic combination should be considered for verification of irreversible serviceability limit states: e.1Qk.i).1 ¦ 2. It corresponds to those effects that are exceeded with a frequency or length of time close to the one of the frequent value \1 Qk.1 ¦ jt1 i !1 2.5 mm crack width is an irreversible limit state. In other words. crack formation).e.i jt1 i !1 Frequent combination: ¦ G k. j P 1.i jt1 i t1 Detailed expressions of the combinations of actions are given in the normative annexes to EN 1990 (Annex A1 for buildings.Guide to EC2 Section 2 If a variable action may be present on the structure at the moment when the accidental action occurs. The characteristic combination should be normally considered for short term limit states. i. and prestressing with its representative value (see ch.i ¸ Qfat i !1 © jt1 ¹ where Q fat is the relevant fatigue load (e.j are taken into account with their characteristic value. Combination of actions for fatigue limit state The combination of actions for fatigue verification is given in [(6.1 of the predominant variable action (e. other variable actions are introduced in the combination with their quasipermanent values (\2.1 will be exceeded. Combinations of actions for seismic design situations ¦G jt1 k. It corresponds to those effects that have a probability of being exceeded which is close to the probability that the characteristic value of the predominant variable action Qk.). SLS combinations There are three combinations of actions for SLS: characteristic. associated with the attainment of a fixed value of these limit states for a long fraction of the reference period (e.i where AEd is the design value of seismic action ( A Ed = I A Ek with JI importance factor and AEk characteristic values of seismic action .3 mm). The frequent combination should be considered for long term limit states.i Qk. the crack width limit state characterized by a 0. j P 1.1Q k. A2 for bridges. Frequent and quasipermanent combinations must be considered for the verification of reversible limit states.2 mm). Note that the seismic action is combined with the quasipermanent value of variable actions.1 Qk.i Qk. the crack width limit state for a reinforced concrete or prestressed concrete beam with unbonded tendons.1 Qk. associated with the attainment of a fixed value of the effect considered for a small fraction of the reference period or for its attainment a fixed number of times.see EN1998).g.i Q k. because such a wide crack cannot completely close once the action that produced it is removed. XC2 exposure class.i Quasipermanent combination: ¦ G k.i Q k. j P A Ed ¦ i t1 2. with design crack width not exceeding 0. etc. Page 212 Table of contents . j P Qk.8.g. associated with the onetime attainment of a fixed value of the effect considered (e.1) will be used.1 ¦ 0. together with the recommended values of partial safety factors JF and of combination factors \.g. with design crack width not exceeding 0. otherwise its quasipermanent value (\2.g.1) will be used in the combination. whereas permanent actions Gk.
10a)1. Sets of partial factors to be used for ULS Limit state EQU – Static equilibrium STR – Design of structural members not involving geotechnical actions STR – Design of structural members involving geotechnical actions (foundations.) GEO – Breaking or excessive deformation of ground (*) Set of partial factors to be used Set A Set B Approach 1(*): Apply in separate calculations design values from Set B and Set C to all actions.00 1. given in Tables [A1.10 0. etc. piles.6.6.00 1. The choice is made in the National Annex.10 or alternatively the most unfavourable between eq.10b is used.35 1.5 1. In general.5 · \L EN1990 (eq.10b)0.16. Sets A.90 1.inf . B and C). Table 2.5 · \ 1.85 · 1. Depending on the limit state under consideration. which are combined in the following table.30 1.5 · \L EN1990 1. Eurocodes allow combinations of actions to contain two or more variable actions.1 (Note 2) Non predominant variable actions Qk. Note 2: The partial factor of favourable variable actions should be taken as 0.00 1.15. In common cases.1 Combinations of actions for the ultimate limit states verification of a building EN1990 Annex A1 gives rules for combinations of actions for buildings.i (Note 2) Set B (Note 1) Set C 1.2(A)EN1990] to [A1.00 1. in the same calculation.5 1.1 G jt1 kj.00 1.sup 0.6. the sizing of foundations is governed by Set C and the structural resistance is governed by Approach 2: Apply Set B to all actions Approach 3: Simultaneously apply.30 Note 1: eq.Guide to EC2 Section 2 2.10)1. on the basis of symbolic expressions and recommended values (or of values given in the National Annex) of partial factors to be applied to actions in the combinations.35 1.6. values of partial factors are subdivided in three sets (A. Combinations obtained with sets A. B and C of partial factors with EN 1990 recommended values are given below. basement walls.2(C)EN1990].5 1. as indicated in the following table. Combinations of actions with Set A of partial factors (EQU) ¦ 1. Set B to actions on the structure and Set C to geotechnical actions The choice of approach to be used for STR/GEO verification is given in the National Annex. values from one or more sets should be used.10a and eq. Note that the partial factor of variable actions should be taken as 0 where these actions are favourable.10a and 6.9 G kj.5 · \L (eq. for ultimate limit states.6.5 · \L EN1990 or alternatively the most unfavourable between the two following: (eq. Table 2.6. the National annex may in addition modify 6.35 1.10b.10a to include permanent actions only. In case of 6.6. B and C of partial factors J for actions Actions Set A Permanent actions Gk Unfavourable Favourable Predominant variable action Qk.
5 i !1 0. 1.1. in the verification of holding down devices for the uplift of bearings of a continuous beam.i In this combination the favourable part of a same action is multiplied by 0. For example.9 whereas the self weight of spans that give destabilising effect should be multiplied by 1. the self weight of spans that give a stabilising effect should be multiplied by 0.i Q k.1 (see Example 2.1).9 and the unfavourable by 1.5 Qk. Page 213 Table of contents .1 ¦1.
sup 1. 0 G kj.inf .Guide to EC2 Section 2 Combinations of actions with Set B of partial factors (STR/GEO) Either ¦ 1.35 G jt1 kj.
sup 1. (6.10)EN1990] (where Gkj.5 Q k.sup are unfavourable permanent loads and Gkj.5 i !1 0.1 ¦1.inf . 1.15 G jt1 jt1 kj. 0 G kj.i [eq.35 G ¦ 1.i Qk.inf are favourable permanent loads) or the less favourable of the two following expressions: ¦ 1.
5 i !1 0.10b)EN1990] kj.1 ¦1.sup 1. (6.10a)EN1990] [eq.inf .5 01 Q k.i [eq.i Qk. 1.i i !1 0. 0 G kj. (6.
For example.sup (Gkj. 0 G 2. Therefore in the verifications of resistance of the sections of a beam. all actions originating from the self weight of the structure may be considered as coming from one source. 1. whereas a different value of partial factor (JG = 1. its self weight should be taken with the same design value for the whole length of the beam ( G = 1.4 Verification of static equilibrium . this also applies if different materials are involved.5 Design assisted by testing 2.5 Q k. [(6.inf .i The National Annex decides whether eq.4. In these expressions Gkj.6 Supplementary requirements for foundations 2.EQU 2. 0 G kj.10a)EN1990] and [(6.1 ¦1. Combinations of actions with Set C of partial factors (STR/GEO) ¦ 1.10b)EN1990] should be considered.inf) is a set of permanent actions from one source with an unfavourable (or favourable) resulting effect of the total action.5 Qk.7 Requirements for fastenings jt1 kj.0) can be taken for permanent loads originating from a different source.35).sup 1.10)EN1990] or the less favourable of [(6.
1,3 Q k,1 ¦1,3
i !1
0,i
Qk,i
See examples 2.1, 2.2, 2.3, 2.4, 2.5
Page 214
Table of contents
Guide to EC2
Section 3
SECTION 3 MATERIALS
SECTION 3. MATERIALS Section 3 of Eurocode 2, dedicated to materials, is structured in the following paragraphs: 3.1 – Concrete 3.2 – Reinforcing steel. Annex CEC2 is related to this paragraph 3.3 – Prestressing steel
3.1 Concrete 3.1.1 General
C3.1.1 General This section deals with normal weight concrete, viz. according to EN 2061 having density greater than 2000 but not exceeding 2600 kg/m3. “Lightweight” concrete is dealt with in Sect.11–EC2.
3.1.2 Strength
C3.1.2 Strength Compressive strength is defined in [3.1.2(1)PEC2], in accordance with EN 2061, by the characteristic value fck (5% fractile of distribution) obtained through the elaboration of compression tests executed at 28 days on cylindrical specimens of diameter 150 mm and height 150 mm. As in many countries testing is carried out on 150 mm cubic specimens, EN 2061 admits fck,cube compressive strength, too. Compressive strength classes are denoted by letter C followed by two numbers that indicate the cylinder and cube characteristic strength, expressed in N/mm2, for example C30/37. EC2 contemplates 14 classes: from C12/15 to C90/105. Table [3.1EC2] gives the numeric values of strength and deformation characteristics associated with strength classes and the analytic relationship expressing such values in function of fck. Average values of compressive strength fcm , of tensile strength fctm and of elasticity modulus Ecm are plotted in Fig. 3.1 in function of fck. Ecm is denoted by the inclination of the line secant of the  relation between points = 0 and = 0,4·fcm as indicated in [Fig. 3.2EC2]. Clause [3.1.2(6)EC2] deals with the development of compressive strength with time. Formula [(3.2)EC2] allows to calculate the average strength fcm at a time t (days) in function of the value at 28 days, which can be deduced from Table [3.1EC2] and from the class of cement used. Cement classes conforming to EN 197 are: Class R (rapid hardening), including CEM 42,5R, CEM 52,5N and CEM 52,5R Class N (normal hardening) including CEM 42,5N and CEM 32,5R Class S (slow hardening) including CEM 32,5N. Table 2.1 shows the development until 360 days of average strength fcm of concrete produced with cement from the three classes, where the compressive strength at 28 days for each class, fcm, equals 1.
Figure 3.1. Values of fcm , fctm and Ecm in function of fck Table 3.1. Development of fcm(t) / fcm,28 ratio t (days) 1 2 3 7 14 28 90 360 Class R 0,42 0,58 0,66 0,82 0,92 1,00 1,09 1,15 Class N 0,34 0,50 0,60 0,78 0,90 1,00 1,11 1,19 Class S 0,19 0,35 0,46 0,68 0,85 1,00 1,18 1,31
Page 31
Table of contents
Guide to EC2
Section 3
3.1.3 Elastic deformation
C3.1.3 Elastic deformation Clause [3.1.3EC2] deals with the elastic deformation represented by the values of the modulus of elasticity Ecm . Note that the numerical values in Table [3.1EC2] and here given in Fig 3.1 are referred to concrete produced with siliceous aggregate, at 28 days of curing. These values should be reduced, for concrete with limestone and sandstone aggregates, respectively by 10% and by 30%. For basaltic aggregates they should be increased by 10%. The development of the Emodulus with time is deduced from the one of fcm (Table 3.1) with formula [(3.5)EC2]. Table 3.2 shows the development of Ecm up to 360 days as a ratio of Ecm at 28 days.
Table 3.2. Development of modulus of elasticity Ecm with time t (days) Class R Class N Class S 3 0,88 0,86 0,80 7 0,94 0,93 0,89 14 0,97 0,96 0,95 28 1,00 1,00 1,00 90 1,02 1,03 1,05 360 1,04 1,05 1,08
3.1.4 Creep and shrinkage
C3.1.4 Creep and shrinkage Clause [3.1.4EC2] is about creep and shrinkage. These two phenomena are typical of concrete. The first relates to the increase in the deformation with time in presence of permanent actions, the second is a spontaneous variation of volume. The development of both phenomena depends on the ambient humidity, the dimensions of the element and the composition of the concrete. Creep is also influenced by the maturity of the concrete when the load is first applied and depends on the duration and magnitude of the loading. C3.1.4 (15) The creep deformation of concrete cc(f,to) at time t = for a constant compressive stress c applied at the concrete age t0, is given by:
3.1.4 (15)
§V · H cc f, t 0
M f. t 0 .
4 (6) C3. t0) is the creep coefficient related to Ec . Annex B of the Eurocode gives detailed information on the development of creep with time.6)EC2] (2.7)EC2]. Where great accuracy is not required. of the notional size ho = 2Ac/u where Ac is the concrete crosssectional area and u is the perimeter of that part which is exposed to drying. N. at the time of loading. the application of an external force and restraint” [JCI.1. in comparison to drying shrinkage. the value found from Figure 3. Table of contents Page 32 . in concrete having a w/c ratio greater than 0. 3.2(6)EC2].4 (6) The total shrinkage strain cs is composed of two components: cd. S) of cement used.1. the autogenous shrinkage strain.1EC2]. In such cases the nonlinear notional creep coefficient should be obtained as from the exponential expression [(3. Moreover.g.45 fck(t0) then the proportionality expressed by [(3. since its value increases with decreasing water cement ratio.45fck(t0) at an age t0. expressed in days. In high strength concrete there is a considerable strength development during the first days.1)EC2] does not subsist and creep nonlinearity should be considered. 1998]. therefore autogenous shrinkage specially has to be regarded in cases that imposed deformations can occur. which develops slowly. Its development in time is linked to the hardening process of the concrete. which may be taken as 1. They are valid up to a concrete strength class C90. In Annex B of EC2 the basic equations for both drying shrinkage and autogenous shrinkage are given. ¨ c ¸ © Ec ¹ [(3. Autogenous shrinkage does not include the volume change due to loss or ingress of substances.30. When the compressive stress of concrete at an age t0 exceeds the value 0. but it can represent 50% of the total shrinkage when w/c is 0.1 are valid for ambient temperatures between 40°C and +40°C and a mean relative humidity between RH = 40% and RH = 100%. provided that the concrete is not subjected to a compressive stress greater than 0. which develops during hardening of the concrete: the major part of it therefore develops in the early days after casting. Autogenous shrinkage is negligible. Autogenous shrinkage specially has to be regarded for higher strength concrete’s. the tangent modulus.45. The values given in Figure 3. the drying shrinkage strain. as detailed at clause [3. graphs are in function of the concrete t0.05 Ecm as from Table [3.1. Autogenous shrinkage can be defined as “the macroscopic volume reduction of cementitious materials when cement hydrates after initial setting. They are also in function of the concrete class (e. temperature variation. C30/37) and of the class (R.1) where M(f.1 may be considered as the creep coefficient. since it is a function of the migration of the water through the hardened concrete and ca. such as in the case that new concrete is cast against old concrete.
Further information is given in Annex B (part B2).5 Stressstrain relation for nonlinear structural analysis Clause [3.1.2 .5 Stressstrain relation for nonlinear structural analysis C3.1. Clause [3. Development of shrinkage according to EC2 3.2EC2] and by the expression [(3.4(6)EC2] gives formulae and tabled values normally used. Shrinkage according to EN 199211 42. In the ENV199211 the following relation has been used in order to describe the mean stress strain relation: NK .1.Guide to EC2 Section 3 Drying shrinkage is essentially a function of the ambient humidity and of the notional size ho = 2Ac/u.2.RH=65% h0=150mm 0 100 200 300 400 500 600 700 800 0 10 C12/15 C20/25 C25/30 C30/37 C35/45 C40/50 C45/55 C50/60 C70/85 C90/105 10 1 Shrinkage [µS] C90/105 C12/15 10 2 10 Age [days] 3 10 4 10 5 10 6 Figure 3.K2 c (1) = fcm 1+ N .14)EC2].5R tcure=1d .1.5EC2] gives the stressstrain relation for nonlinear structural analysis as described by [Fig. 3.
0022 c1/fc (strain at peak compressive stress) (3) k = (1. K where = c / c1 (2) c1‰ = 0. stress in vertical axis in MPa) Page 33 Table of contents .1 Ec) Ec denotes the mean value Ecm of the longitudinal modulus of deformation. Fig. 3.2 shows compressive stress–strain relations for concrete strength classes ranging from about C25 to C90 Figure 3. however. High strength concrete is known to behave in a more brittle way and the formulation therefore cannot be extended to high strength concrete without modification.2 Stressstrain relation for concrete’s different strength classes subjected to a constant strain rate (strain in horizontal axis in ‰. where Ecm = (fck + 8)1/3 (4) In ENV 199211. only concrete strength classes up to C50/60 were considered.
Page 34 Table of contents .221 A warning is given that the descending branch highly depends on the testing procedure and its formulation should be used with caution.3.070 fck (MPa) Hc0 (103) 50 0.3. a simplified formulation is preferred. 1.338 0.8 + 27 [(98 – fcm)/100]4 for fck 55 Mpa (8) In this way the simplified curves shown in Fig. testing of the concrete considered is recommended). Eq.7) the diagram shown in Fig.0. using Eq. 3. This equation is also a good approximation for the Emodulus of normal strength concrete and could therefore be attributed general validity.5.1) / ( 2 – 1)}2] 1 (7) c0) where = c / c1 and 2 =( c1 + / c1 where c0 is a value to be taken from Table 3.Guide to EC2 Section 3 In [CEB.807 Table 3.2.0022 should for HSC be replaced by c1 (‰)= .000 [(fck + f) / 10]0. 90 0.The parameter t for HSC 60 70 80 0. the constant value c1 = .3 is obtained. according to cu ( ‰) = 2. the following modifications have been proposed: Eq. 6. (4) overestimates the Emodulus for HSC. in that the lines according to Eq.3. 1995]. To determine the ascending branch.579 0. obtained by combining Eq.3 (5) where the difference between mean and characteristic strength f is 8 MPa. An appropriate formulation for HSC is: Ecm = 22. Using the relations (1.31 (fcm in MPa) (6) This value can as well be used for normal strength concrete For HSC (>C50) the descending branch should be formulated by c = fcm / [1 + {( 1 . (It should be noted that the given values are mean values and that the real modulus of elasticity can considerably be influenced by a component like the aggregate. Figure 3. (1) are continued beyond the top.4 may be obtained.0.3.7 fcm0. If the modulus of elasticity is important and results from similar types of concrete are not known. Mean stressstrain relations. 3. (13) and (57).6. Since the descending branch for HSC is not very reliable. until a defined value cu is reached.
1960]. Figure 3.5 reproduces Rüsch’s diagrammatic representation of concrete strains as a function of the applied stresses for several loading times. the longer the loading time. 1960) As can be seen. which he loaded to a certain fraction of the shortterm compressive strength: subsequently the load was kept constant for a long period. If the longterm loads were higher than about 80% of the shortterm bearing capacity. the more the ultimate strength approaches the longterm value 80%.5.2) where cc is the coefficient taking account of long term effects on the compressive strength and of unfavourable effects resulting from the way the load is applied. failure occurred after a certain period.6 Design compressive and tensile strengths The value of the design compressive strength fcd is defined as fcd = Dcc fck /JC [(3. Fig. Page 35 Table of contents . according to the new formulation. which is 1. The tests carried out by Rüsch were limited to concrete’s with a maximum cube strength of about 60Mpa. Stressstrain relations for several time durations of axial compressive loads (Rüsch. (5.4.1NEC2]. Tests by Walraven and Han on concrete’s with cube strength’s up to 100 Mpa showed that the sustained loading behaviour for high strength concrete is similar to that of conventional concrete’s [Han/Walraven. 1993].1. combining Eq. Simplified mean stressstrain relations.1.15)EC2] (2.6 Design compressive and tensile strengths C3.Guide to EC2 Section 3 Figure 3. 3. A well known research program focussing on the effects of long term loading was the one carried out by Rüsch [Rüsch. He carried out tests on concrete prisms.8) 3.50 [Table 2. C is the partial safety factor for concrete.6.
which generally will be much older when subjected to a load. In order to obtain consistent and sufficiently safe design relations the ultimate strains cu have also been slightly reduced in relation to Eq. Compressive strength development of concrete made with various types of cement according to Eq. Fig. This can principally be done using the relations given in 3.5 it can be seen that in a test with a loading duration of 100 minutes. unless specified otherwise”. Therefore in clause 3. Page 36 Table of contents . Figure 3.1.5. all curves end approximately at their top (compare Fig. 3.6 + 35 [(90 – fck)/100]4 < 3. fractile 5%.1.3 for concrete’s made with rapid hardening high strength cements RS. normal and rapid hardening cements N and R.5 (9) for concrete strength classes C55 and higher. but now for the characteristic compressive strength (fck) values and subsequently reducing the stress ordinate by a factor Jc = 1.Guide to EC2 Section 3 However.16)EC2] (2. The value of the design tensile strength fctd is defined as fctd = Dct fctk 0.3).6 shows the strength development in time according to eq. A certain sustained loading effect is therefore already included in the results of tests.05 /JC [(3. 8. In Fig.5 hours.6 shows.2. This means that the sustained loading effect is at least partially compensated by the increase in strength between 28 days and the age of loading.1.6.1EC2]. and slowly hardening cements SL. that the gain in strength in 6 months ranges from 12% for rapid hardening cements to 25% for slowly hardening cements. torsion.4 it is stated that “the value of cc may be assumed to be 1. EC33 Fig 3. capacity of columns). similar to cc.1. The resulting curves are shown in Fig. This condition will normally not hold for a structure in practice.8.7 Stressstrain relations for the design of crosssections C3. punching. EC3. For tension similar arguments apply. 3. the reduction of strength with regard to 2 minutes is already about 15%. By choosing the expression cu (‰) = 2. which can be deducted from Table [3. 3.05 is the characteristic axial tensile strength of concrete. 3. a considerable part of the sustained loading effect is compensated for by the increase in strength.7 Stressstrain relations for the design of crosssections Design stressstrain relations can be derived from the mean stress strain relations. Furthermore the bearing capacity as formulated in building codes is generally based on experiments in laboratories (shear. Such a case can for instance occur when. It is therefore concluded that cases in which the sustained loading effect will really influence the bearing capacity of a structure in practice are seldom and do not justify a general reduction of the design strength with a sustained loading factor of 0. Normally those tests have a duration of at least 1.1. This means that principally not only the stress values but also the Ec values are divided by Jc.5. 3. fctk 0. according to 3. the concrete strength is determined substantially after 28 days: in such a case the gain in strength may be marginal so that a value cc smaller than 1 is more appropriate. Rüsch’s tests were carried out on concrete which had an age of 28 days at the time the load was applied. the value of which may be found in the National Annex.3) where ct is a safety factor. So.7.
4 C90 2.8) is expressed by Vc Vc where n.5 n 2 C35 2.1 1.3 2. Parabolarectangle design stressstrain relation Figure 3. and one on the basis of a bilinear relation (Fig.9.7 1.6 1.45 C80 2.6 C70 2. Design stressstrain curves for a parabolarectangle formation The parabola rectangle relation (Fig. Parameters for the parabolarectangle design stressstrain relation in compression C20 Hc2(‰) 2.0 Hcu2(‰) 3.75 C60 2.5 2 C55 2.4 2.4.4 Page 37 Table of contents .Guide to EC2 Section 3 Figure 3.5 2 C50 2.2 3.0 3.5 2. 3.9 1.6 1. 3.7.9) Figure 3. ª § H ·n º fcd «1 ¨ 1 c ¸ » « © Hcu ¹ » ¬ ¼ fcd c2 for 0 d Hc d Hc 2 for 0 d Hc d Hc 2u and c2u follow from the table below: Table 3.0 3.6 2. Basis for design diagrams In the new version for EC2 two alternative design curves are given: one on the basis of a parabolarectangle relation (Fig.8. 3.8).
Rectangular stress distribution As a basis for the derivation of O and the parabolicrectangle stress strain relation is used.12. This could be improved by reducing the values c3 for the lower strength classes.12). This requires the introduction of a factor for the depth of the compression zone and a factor for the design strength. Figure 3.14. Derivation of rectangular stress block from the parabolicrectangle stress distribution for concrete strength class C50 Page 39 Table of contents . Since the loss of accuracy for practical calculations is very small the constant value c3 = 1. Comparison of “basic” and “approximate” design curves In clause 3.75 % should be maintained. The figure shows that there is good overall consistency. large sets of design diagrams for various cases. As an example the values O and are calculated for the strength classes C50 and lower. Figure 3. Now a rectangular stress block is searched for.13. It could be argued that for the lowest concrete strength classes (see the figure for C20 in Fig. For concrete’s in the strength classes C50 the characteristic strains are c2 = 2.0 ‰ and c2u = 3.1. see fig.11. see Fig. 12. 3. the bilinear approach is slightly less accurate. which gives the same resulting force at the same location. 3. This would however have a considerable disadvantage since then the shape of the design stress strain curves for all strength classes < C55 would be different which would render substantial complications for practical design (different shape factors and distance factors. Figure 3. where now only one diagram is sufficient). 3.5‰.13. see Fig.6 (4) of EC2 as well the possibility is offered to work with a rectangular stress distribution.Guide to EC2 Section 3 on the basis of the parabola – rectangle approach and the bilinear approach.
Guide to EC2
Section 3
For the parabolicrectangle stress distribution the resulting force is 0.81 xbfcd and the distance of this force to the top is 0.415x, where x is the height of the compression area. In order to obtain a rectangular stress block with its resultant at the same position, the depth of the compression area should be x = 2*0.415x = 0.83x, so = 0.83. In order to get the same magnitude of the resultant, the maximum stress is defined as fcd. The resultant force for the rectangular stress block is ( x)b( fcd). Since this force should be equal to 0.83xbfcd, the value of follows from = 0.81/ = 0.98 Carrying out this calculation for all concrete strength classes, with the values for c3 and c3u taken from table 3.5 [table 3.1 EC2], the values and O shown in fig. 3.15 are obtained.
Figure 3.15. Values and for the definition of the rectangular stress block
Approximate equations for and are: O 0,8
O 0,8 
per fck 6 50 N/mm2 per 50 < fck 6 70 N/mm2 per fck 6 50 N/mm2
fck 50
400
K 1
K 1,0 
fck  50
200
per 50 < fck 6 70 N/mm2
If the width of the compression zone decreases in the direction of the extreme compression fibre, the values given in Eq. 16a,b do not hold. This is investigated in Fig. 3.16.
Figure 3.16. Derivation of rectangular stress block from bilinear stress block for a cross section width in the direction of the extreme compression fibre
Similar type of calculations has been carried out for this case. Fig. 3.17 shows for an as a function of the concrete strength. Three cases are considered:  the calculated relations for a rectangular crosssection  the calculated relations for a triangular crosssection with the top at the extreme compressive strain  the design equation, derived for the case of the rectangular crosssection. It is shown that the design equation for is safe for both cases. However, the values for for the triangular case are lower than the design values. It is sufficient to reduce the values for with 10% in order to cover as well the triangular case.
Page 310
Table of contents
8 Flexural tensile strength The flexural tensile strength is larger than the concentric tensile strength.2(3P)EC2] states that the design rules of Eurocode are valid when steel having characteristic yield fyk between 400 and 600 N/mm2 is used.9 Confined concrete 2 at the ULS as a result of confinement. temperature gradients and imposed deformations. the axial 0.2.c is: fck.2. 1995). 3.9 Confined concrete Having a uniform radial compressive stresses strength fck.8 Flexural tensile strength C3. in this case the phenomenon is more sensitive to the effect of drying shrinkage.c = fck [1.2.4) (2.000 ( 2 /fck)] fck.1.5 Welding 3. However. This is caused by strain softening at cracking concrete.2.5) for for 2 2 Axial strength increases by 50% if the lateral compressive stresses are 15% of fck.25)EC2] (2.fl ¨ 1.6 Fatigue Page 311 Table of contents .5 ( 2 /fck)] 3. Calculated values for and for a rectangular and triangular crosssection.1.c = fck [1+5. Also the strain at failure increase up to 34 times.17.05 fck [(3.1. 3.24)EC2] [(3. so that the size 200 factor k 1 may be expected to apply for the relation between flexural tensile strength and d concentric tensile strength as well. Therefore the more conservative equation h · § fctm . The size effect of concrete in bending is therefore the same as for concrete subjected to shear or punching (see f.1. Lateral stresses may be achieved by confinement of the compressed member. stands for "confined". Index c after fck .1 General 3.3 Strength 3.05 fck > 0.2 Properties Clause [3.Guide to EC2 Section 3 Figure 3. C3.2 Properties C3. in comparison with the design equation for the rectangular case.2. 3.6 ¸ fctm t fctm 100 ¹ © has been chosen.2.i.125 + 2. Walraven. it doubles if compressive stresses are 35% of fck.2.4 Ductility characteristics 3.2.2 Reinforcing steel 3.
Guide to EC2 Section 3 3.3 Prestressing steel 3. ¨f ¸ © y ¹k The value of the elasticity modulus Es may be taken as 200000 Nmm2 .3.2.3 Strength Within each type.9 uk. 3. is between 1.8EC2] For normal design.18.1NEC2]. Relevant criteria and methods for verification are 3. The design stressstrain diagram is shown in Fig. Although it's not indicated in EC2.2 Properties wires with plain or indented surface. 3. Clause [3. nominal diameter of the strand between 5.5 and 5. Ductility.5% for Class 2.18.35.3. of diameter between 3. Figure 3.15 according to Table [2.1k).3 Prestressing steel Data from EC2. denoting the value of the ratio of tensile strength to proof strength (fpk /fp0.2 and 7. 2. Three classes are defined in the Eurocode: • Class 1: wire or strand – ordinary relaxation • Class 2: wire or strand – low relaxation • Class 3: hot rolled and processed bars The design calculations for the losses due to relaxation of the prestressing steel should be based on the value of 1000.1k).7 mm sevenwire strands of which a straight core wire around which are spun six wires in one layer. given in Annex C for class C steel. nominal diameter of the strand between 4. and elongation at maximum load ( uk).0 mm ribbed bars. nominal diameter of the strand between 6.0 and 50.0 mm. indicating the relaxation behaviour.1 General C3.035.0 and 11.7 Design assumptions C3. Idealised and design stressstrain diagrams for reinforcing steel [Fig.4 Ductility characteristics Page 312 Table of contents . 1000 values indicated in EC2 for structural design are: 8% for Class 1. where fp is the actual tensile strength of the prestressing steel samples. which should be at least 1.6 mm threewire strands spun together over a theoretical common axis.2.3. either of the following assumptions may be made: c) an inclined top branch with a strain limit of ud and a maximum stress of k §f where k = ¨ t ¨f JS © y fyk · ¸ ¸ ¹k d) a horizontal top branch without the need to check the strain limit.7fp. integrated with those from EN 10138 which is referred to in EC2.15 and 1.7 Design assumptions Safety factor JS for ultimate limit states is 1.4 and 18.3. Fatigue: prestressing tendons are liable to fatigue.1. denoting the value of tensile strength fp and the value of the 0. Prestressing steel are geometrically classified as: 3. The recommended value of ud is 0. 3. The value of 1000 is expressed as a percentage ratio of the initial stress and is obtained for an initial stress equal to 0. 3.2(7)EC2] gives the formulae for calculation of relaxation at different t times for the three abovementioned classes.1% proof stress (fp0. reinforcing steel is classified according to the following properties: Strength. the relaxation loss (in %) at 1000 hours after tensioning and at a mean temperature of 20 °C. §f · The value of ¨ t ¸ . Class.3. 4% for Class 3.0 mm twowire strands spun together over a theoretical common axis. are recalled hereafter. nominal diameter between 15. in accordance with EN10138 uk should be at least 0. Annex [DEC2] provides the elements needed for accurate calculations.
A.4 Prestressing devices LITERATURE Aitcin.”Second Stichtse Bridge – Concrete with high strength: experimental research into the creep and relaxation behaviour during the early stage of the hardening process”. 3. Delft University of Technology. ”Creep of High Performance Concrete – Characterisation and Code Type Prediction Model”. V.. “Creep and shrinkage models for normal and high performance concrete – a unified code type approach”. 4th International Symposium on Utilisation of Highstrength/ High Performance Concrete.3. E. “Time dependant behaviour of high strength concrete”. pp. Feb. Report 25. C. Kvitsel.Guide to EC2 Section 3 3. TU Delft (in Dutch).5 Fatigue given at clause [6. van Breugel. JCI Technical Committee Report on Autogenous Shrinkage Autogenous Shrinkage of Concrete: Proceedings of the International Workshop.9 uk. Küttner.19. As represented in Fig. Revue Francaise du Genie Civil (1999) Neville.3. Tazawa. 1992 fib. H. On top of strength and ductility values.10EC2] 3.S.5962. pp... PhD thesis. 5456.. July 1995. the design diagram is made of a rectilinear part up to ordinate fpd from which two ways start: a rectilinear inclined branch. Design and Performance. K. taken the safety factor S = 1.8EC2]. London 1995. Concrete International. Upgraded Knowledge of the CEB/FIP Model Code Han.. Koenders. 1999.. Pitman Publishing Limited. or 0. “Structural Concrete: Text Book on Behaviour. CEBBulletin 228. the other branch is a horizontal branch without strain limit. P. the Eurocode provides the following design assumptions: Modulus of elasticity recommended for strands: 195000 N/mm2.H. “Demystifying Autogenous Shrinkage”. 1998. Müller.. The Netherlands”.15. 3. “Properties of concrete”.S.M. specifications and conformity criteria – Part 1 Common cements.. N. 377/386 Müller. ed. 3. 1996. Paris.. Page 313 Table of contents . Küttner. London. E&FN Spon. “High Performance Concrete: Recommended Extensions to Model Code 90”.02. H. with a strain limit ud = 0. 1996. ENV 1971 Cement – composition.H..6 Design assumptions Figure 3. pp. 1996.7 Prestressing tendons in sheaths 3. 363. for wires and bars: 205000 N/mm2 Design stressstrain diagrams.3.19. Idealised and design stressstrain diagrams for prestressing steel [Fig. Report on the CEBFIP Working Group on High Strength Concrete”. C.
0 n = 1.75 3. (69) The values for c2 and n are obtained by curve fitting to the relations shown in Fig.10). Design stressstrain curves with bilinear formation Fig.11. Figure 3.5) and the “simplified” design curves.0 + 0.4 the values for c2u follow from Eq.7 C80 2.5 C55 1.9 C70 2.11 (fck – 55)0.5. 3.6 C90 2. 3.5.4 + 23.10.4 [(90 – fck)/100]4 for fck 50 Mpa for fck > 50 MPa (13a) (13b) The resulting design curves are shown in Fig.6 The values for c3u are the same as for expression for c3 is c3 and follow therefore from Eq. 3. A second possibility is the use of a bilinear design stressstrain relation (Fig.0 for fck 50 Mpa for fck > 50 Mpa (12a) (12b) c2 (‰) = 2. 3. 3. 3.75 3.5 C50 1.085 (fck – 50)0. An approximate (‰) = 1.5 C35 1.3 2. Bilinear design stressstrain relation Fig. Table 3.75 3. An approximate expression for c2 is: c2(‰) = 2.7.42 for fck 55 MPa (14a) (14b) c3 The design stressstrain relations derived in this way are shown in Fig.11.Guide to EC2 Section 3 In Table 3.1 2.75 for fck < 55 MPa (‰) = 1.75 Hcu3(‰) 3.10. Parameters for the bilinear design stressstrain relation in compression C20 Hc3(‰) 1. Page 38 Table of contents .3.53 and for n: n = 2.75 + 0. and are given in Table 3. 9. The values c3 and c3u are obtained by curve fitting to the relations shown in Fig.1 c2u C60 1.12 shows the comparison for the “basic” design stress – strain relations (derived from the mean curve by taking the characteristic value and dividing it by Jc = 1.90 2.7.2 2.
Walraven. pp 114. Stavanger 1987. Proceedings of the Conference Utilisation of High Strength Concrete”. “Size effects: their nature and their recognition in building codes”. 113134. Calgary. Walraven. J. Proceedings. J. a European Perspective”. Page 314 Table of contents . “High Strength Concrete.. “High Performance Concrete Bridges”.An Overview of Concrete Research”. STUDI E RICERCHE – Vol. Part I... ’98. 16. A.Guide to EC2 Section 3 Nilson. Canada. Conference on Developments in Short and Medium Span Bridge Engineering.C.C. 1995. pp.
1. This rule was changed with respect to the general requirement that values or rules specified in other Eurocodes should not explicitly given again in EN 199211 but they should only be referred to. Previously the concrete cover was prescribed in dependence of the environmental class. based in the model in the European project.1): xc t . In the following the timedependent description of the carbonation progress and the time dependent diffusioncontrolled penetration of chlorides are briefly presented. According to the first draft of prEN 19921 the allowance in design for tolerance was also dependent on the environmental class – 10 mm for XC0 and XC1 and 15 mm for all other classes. 34 “Model code for service life design” [2].1.1 CarbonationInduced Corrosion The CEB Task Group V model by which the carbonation process in the initiation period can be predicted is given in equation (4. In the following. cracking or spalling of the concrete cover will occur.2. are well established. In the literature deterioration models [2]. 4. by which the processes of the initiation period can be described. As a result of corrosion a reduction of the steel cross section.2. After depassivation. which consists of the minimum concrete cover (dependent on the relevant environmental class) plus an allowance in design for tolerance. This means a reduction of the nominal cover if the values for the minimum cover are not increased accordingly. In order to find out which value for the nominal concrete cover is adequate a durability design was performed. DURABILITY AND CONCRETE COVER C4 The rules on design for durability in EC2 are substantially different than in the past. 4. In the following theoretical considerations the most important deterioration processes are explained and parameter studies illustrate the mutual dependencies. since the deterioration models related to reinforcement corrosion and the safety concept of durability design have already been thoroughly described in the literature. In the actual version of EC2 (EN 199211) the cover required depends not only on the environmental class. and practical experience.1 Introduction In a European research project [1] a probabilistic based durability design procedure of concrete structures has been developed with the objective to set up a similar concept as in structural design.1 Deterioration Models The corrosion process can be divided into two time periods: the initiation period describes the time until the reinforcement is depassivated either by carbonation or by penetrating chlorides reaching a critical chloride content. background information is given with regard to those choices. In particular. as revealed by the Duracrete studies [1].1. Related to corrosion protection of reinforcement the resistance of a concrete structure is mainly determined by the thickness and the quality of the concrete cover. Further background information is found in FIB Bulletin n. Thereafter in the December 99 draft of prEN 19921 the allowance in design for tolerance was determined with respect to the execution standard prENV 13670 where the execution tolerance is uniformly defined to 10 mm. This time period is described as the propagation period. the required design working life and the quality control applied.2 Description of Deterioration Models and Probabilistic Durability Design This chapter only gives a short overview on those models. In the European design code for concrete structures EN 199211 the designer has to determine the nominal concrete cover.1. The concrete mixes have been chosen with respect to the environmental classes given in EN 2061. where the resistance of the structure is compared to the acting load.1.Guide to EC2 Section 4 SECTION 4 DURABILITY AND COVER TO REINFORCEMENT SECTION 4. The values for the cover in EN 199211 are a result of increased understanding in the processes of deterioration. 4. corrosion will start if sufficient oxygen and moisture are available. 4. but as well on the concrete strength class. The process of the propagation period is much more complex and so far no unanimously accepted models exist. the reliability index E was determined and evaluated for different concrete mixes and different nominal concrete covers. but independent of the concrete quality.
= 2 k e k c DEff. humidity) Page 41 Table of contents . 65% rel.1) where: xc(t) DEff.0 is the carbonation depth at time t effective diffusion coefficient of dry concrete for carbon dioxide in defined environment (20°C.0 C a §t · t ¨ 0 ¸ © t ¹ w (4.
describing wetting and drying reference period.1. describing the mean moisture content of concrete parameter to describe the curing conditions parameter (exponent) for micro climatic conditions at the concrete surface. which usually means the carbon dioxide content of the surrounding air c0 parameter for micro climatic conditions. 1 year) t time 4.2): §t · x t . 'C ke kc w t0 t law valid (e.1.2.g.2 ChlorideInduced Corrosion The model for predicting the initiation period in the case of chlorideinduced reinforcement corrosion is defined by equation (4.Guide to EC2 Section 4 a the amount of CO2 for complete carbonation [kgCo2/m³] the concentration difference of CO2 at the carbonation front and in the air.
= 2C crit .
0 k e ¨ 0 ¸ t © t ¹ n (4.3) ktDRCM.0 = D0 § C · C crit . k t DRCM.2) (4.
They are generally solved by facing the two variables R and S: Z=R–S (4. and in this example they are normally distributed. measured at time t0 chloride threshold level factor which takes the influence of age on material property into account constant which transforms the measured chloride migration coefficient DRCM. R the resistance and S the action: both variables R and S have their averages and standard deviations.6) and (4.0 chloride migration coefficient under defined compaction. the reliability index is the difference between the mean values of R and S divided by the standard deviation of the variable Z or. = erf 1 ¨ 1.7).6) (4.2.1) and is also normally distributed with a mean Pz and a standard deviation Vz according to equations (4. Figure 4.crit ¸ © CSN ¹ where: x(t) D0 depth with a corresponding chloride content (here C(crit)) at time (t) (4.1): Page 42 Table of contents . PZ PR PS Vz 2 2 VR VS (4.4) effective chloride diffusion coefficient under defined compaction.1. Z is a variable itself (see Figure 4.8.2.2.7) In this simplest case with two variables.5) where Z is the reliability of the structure.1 Safety Concept The simplest design problems have only one resistance variable and one action variable.0 into a chloride diffusion coefficient D0 constant which considers the influence of environment on D0 inverse of the error function surface chloride level time reference period (28 days) Ccrit n kt ke erf1 CSN t t0 4. measured at time t0 DRCM.2 Probabilistic Durability Design 4. alternatively the mean value of Z divided by the standard deviation of Z (see Equation 4. curing and environmental conditions. curing and environmental conditions.1.
Guide to EC2 Section 4 pf § P · ) ¨ z ¸ ) E .
XD1) or environments with cyclic wetting and drying (XC4. XS1. fly ash). the durability design presented here durability design is a SLS assessment. A possible way of taking account of the different corrosion risks is to adjust the reliability index to the environmental classes.50 for a life time of 50 years 4.3. All input values are given as stochastic variables (mean. As the corrosion process in the propagation period can lead to severe consequences (loss of safety of people and structure). Since the calculations performed in this study only concern the initiation period. The output of the program is the reliability index E as a function of the lifetime. the corrosion process can be divided into two time periods. because after the end of the initiation period the corrosion process starts [1]. limit state conditions can be defined. with respect to the corrosion process a moderate humid environment (e.3. XC3. XD2) and therefore a higher reliability index is proposed. the porosity of the concrete increases after carbonation and the carbonation of the concrete sometimes progresses faster (as the diffusion of CO2 takes place through the carbonated zone). For the same reason in chloride environments higher reliability indices should be applied due to the risk of higher corrosion rates compared to carbonation induced corrosion.2. 4.g.2 Carbonation The type of cement has a strong influence on the concrete resistance against carbonation.5 0.2. XC2.0 1. In this study the reliability index E is considered for a lifetime of 50 years.5 Table 4. action. As a consequence. For this reason. Because of the fact that only the initiation period is considered. Figure 4. this situation is defined as ultimate limit state (ULS).XC1.1: Required reliability index bSLS. type of distribution). XD3.1.1 resistance. variables are not always normally distributed. in other National Standards the reliability index is even higher.2. XD3 XC2. Concretes with Portland cement are exposed to carbonation they show a decreasing porosity in the carbonation zone resulting in higher resistance against ongoing carbonation.2. XS3. In concretes with blastfurnace cements or with cements with pozzolanic additions (e. Environmental class XC4.3 Parametric Study 4. The influence Page 43 Table of contents .5. XS2.1 and 4.) stands for a normal distribution and pf for the failure probability. Design limit states are often defined by means of the reliability index E According to EN1990.g. XS2.g. In the case of durability design a risk oriented grading of the reliability index is proposed. computer programs are used to calculate these kinds of problems.50 2. the corrosion process itself is so far not included in durability design. Depassivation of the reinforcement (= end of the initiation period) is defined as a serviceability limit state (SLS). because there are numerous variables which have to be statistically evaluated and they are often part of nonlinear functions. A proposal is given in Table 4. on contrary. Besides these aspects. standard deviation. Definition of reliability levels As already mentioned. XD2 XC1 Reliability index ESLS. For both periods. failure probability and reliability index The numerical problems which are treated in the context of durability design are not as easy to calculate as this.1.1 General The calculations are based on the models given in 4. 4. © Vz ¹ where )(. XC3. XD1. the reliability index for SLS is determined to E = 1.2. XS3) should fulfil higher safety requirements than totally dry or wet environments (e.
durability design calculations for nominal covers of 30 and 35 mm.3a).3).2: Reliability index versus concrete cover for environmental class XC2 (w/c=0. for different climates and for different cements (CEM I and for CEM III/B) are performed. The length of the curing period also has a strong influence on carbonation.2. However it needs to be taken into account that part of the effect depending by the type of cement can be counteracted by prolonged curing.Guide to EC2 Section 4 of different types of cement is not considered.5 R (curing 2 days) b) for CEM III/B 42.1 Environmental class XC2 In the following. Especially in the case of a CEM III/B.1a. It is interesting to see that the choice of a different type of cement has a bigger influence than the reduction of the concrete cover.3.5 NW HS NA (curing 3 days) It can be seen that the reliability index for the calculated values depends on the concrete cover.5 and compared to the environmental classes XC2 and XC3 the concrete cover is increased by 5 mm. The cement content was kept constant (320 kg/m³) for all concrete mixes.6. In Figure 4. This means a reduction of life time of more than 15 years. Apart from this the same trends can be observed.5 R (curing 2 days) b) for CEM III/B 42.3. Figure 4.2. that this improvement of the concrete quality (porosity) also has an important effect on the carbonation (compare Figures 4. In the calculations the curing time was assumed to be 2 days for CEM I and 3 days for CEM III/B. In the calculations two different concrete mixes (favourable and unfavourable) were studied. but the required w/c ratio for class XC2 is higher. The reduction of the nominal concrete cover from 35 to 30 mm has a big influence on the reliability index. the location and the type of cement. Figure 4. 4. the w/cvalue is reduced to 0. The climatic conditions were determined according to statistical data from local weather stations. The w/c ratio of the concrete mixes were chosen according to the requirements in prEN 2061. because there is only a minor influence on the carbonation process.5 NW HS NA (curing 3 days) 4. The material resistance was determined under laboratory conditions.3 Environmental class XC4 For environmental class XC4. This has a considerable effect on the reliability index (compare Figures 4. 4.2a and 4. service life 50 years) a) for CEM I 42. It can be seen.3. even more pronounced than for concretes with a w/c ratio according to the requirements for XC2. 4. the reduction of the concrete cover by 5 mm results in a decrease of the reliability index of almost 50 %.55.service life 50 years) a) for CEM I 42.2 and 4.3: Reliability index versus concrete cover for environmental class XC3 (w/c=0.0 for XC4 all concrete covers are still above the value for Page 44 Table of contents .1 is increased to 2. Although the required reliability index according to Table 4.2 Environmental class XC3 For environmental class XC3 the same concrete covers are required as for XC2.2.2 the results of the calculations are shown. All calculations were performed for European locations with moderate and hot climate to show the influence of temperature.
45) and the cement content was 320 kg/m³. However when using a CEM III/B type of cement. (see Figure 4. In general. An additional reduction of the concrete cover means a further reduction of the reliability index. the use of Portland cements leads to a higher permeability of the concrete for chlorides than the use of blastfurnace cements or cements with fly ash. The calculation with Portland cement plus fly ash has shown similar trends as in the case of carbonation. Figure 4. In this study only environmental class XS3 with nominal concrete covers of 50 and 55 mm was studied. The reliability index for a lifetime of 50 years was below 0. harbours) has already led to deterioration (cracking) as a result of chloride induced corrosion after 10 years.4 . the reliability index is reduced by 50% and is far below the reliability index proposed for this environmental class.Reliability index versus concrete cover for environmental class XC4(w/c=0. Examples have shown that the use of Portland cements in tidal environments (e. The time before attainment of a critical chloride content depends mainly on the porosity of the concrete. Processes of diffusion and conveyance by water transport take place which lead to a certain chloride content in the concrete. Figure 4. It can also be seen. Alternatively the calculation was performed with CEM I without fly ash [1]. the model seems to put in evidence that the deviation of the proposed reliability index is still acceptable.5 shows that hot climates result in faster chloride ingress than moderate climates.5: Reliability index versus concrete cover for environmental class XS3 (w/c=0. The climatic conditions (relative humidity and temperature) were determined according to statistical data of local weather stations and the calculations were performed for two European locations close to the sea in moderate and hot climate.5 NW HS NA 4.g.50) a) for CEM I 42.3 Chloride penetration Chlorides are transported in the concrete by the pore water. The w/c ratio of the corresponding concrete mix was determined according to EN 2061 (w/c = 0.Guide to EC2 Section 4 CEM I.5 R b) for CEM III/B 42. The calculations were performed with two different types of cement [1]: the results for a CEM I with fly ash are shown in Figure 4.45) a) for CEM I with fly ash Page 45 Table of contents . that means that corrosion probability is higher than 50%. which may be influenced for example by the w/c ratio and the type of cement.3b).3. that the proposed reliability index is not reached for environmental class XS4.4. Figure 4. The considerable difference in reliability index as a consequence of using two types of cements shows that the currently existing requirements for the determination of the concrete cover do not include all important parameters and are therefore not very exact. For a nominal concrete cover of 40 mm. whereas a nominal concrete cover of 35 mm leads to an unacceptable decrease of total life time.
Report No. See example n. Further influences. the thickness of the concrete cover and the environmental class. ISBN 9782883940741 Page 46 Table of contents . 4. that there are many uncertainties in the input values of the calculations. 34 Model Code for Service Life Design. 4. The comparisons show that.4 Conclusions The calculations have shown the dependence of the rate of deterioration on the service life. local differences due to different orientation with regard to solar radiation and wind). Further to the influence of the type of cement and the temperature there is the variation in climatic conditions (wet .2 and 3.3 have shown that lower values for the nominal concrete cover result in a significant reduction of the reliability index. 4. theoretically. Reference [1] DuraCrete – Probabilistic Performance Based Durability Design of Concrete Structures: Statistical Quantification of the Variables in the Limit State Functions. which have not yet been regarded in detail in EN 199211 are the type of cement and the temperature. It should be realized. research is necessary to close the gap between scientific models and practical observations. pp. 4.2.3. 4. Moreover. regarding the many unknown factors. It can be concluded that the nominal concrete cover should not be decreased as a consequence of the reduction of allowance in design for tolerance. the prescribed reliability indexes are not always met. leading to a decrease of the safety against reinforcement corrosion. 6263. The advised increase of 10mm is therefore a good average value. 2000. The recommendations in the code are the result of theoretical considerations and engineering experience.Guide to EC2 Section 4 The results of chapters 3. June 2006.dry cycles. In some respects the deterioration models give valuable information. It is shown for instance that prolonging the service life of a structure from 50 to 100 years requires globally an increase of the concrete cover between 8 and 12 mm. however.4. It is shown that the provisions as given in EC2 recognize the most important influencing factors.: BE 951347. [2] FIB Bullettin n.1.
assumes at ultimate limit state the transformation of the structure in a mechanism through the formation of plastic hinges. The present value is also well correlated with the tolerances given for class 1 in EN 13670. The analysis implies a preliminary idealisation of the structure. plastification of reinforcement steel beyond yielding. However.plastic behaviour.2 Imperfections and tolerances The minimum value of the basic inclination is now 1/300 instead of 1/200 and 1/400 (with and without 2nd order effects respectively). . The fact that the mean imperfection i is sometimes less than the structural tolerance does not Page 51 Table of contents .4EC2]. The thin lines represent the tolerance according to EN13670. based on more or less refined assumptions of behaviour.2.Guide to EC2 Section 5 SECTION 5 STRUCTURAL ANALYSIS 5.7EC2]. The rules in the EN are technically rather similar to those in the ENV.2. STRUCTURAL ANALYSIS C5. It is a design (not analysis) procedure based on mixed assumptions. the discontinuities mentioned above have been removed. . There are four types of idealisations: . Geometric imperfection 5.6EC2]. and the rules have been coordinated between EC2. that takes into account. the dashed line the total inclination of a structure. in its static approach. The design procedures for nonlinear analysis are given in [5. The mean value of the imperfection will also decrease in a structure where a number of individual members contribute to the total effect. Its kinematic approach [5.5. the lower limit of the mean value (for m = ) is shown in figure 5. The thick lines represent imperfections. impressed strain) in terms of tensional states or strain on a geometrically and mechanically defined structure.1 General SECTION 5. see figure 1.2.2.1. whereas the grey line represents the lower limit of the mean value Tl for a structure (m = ).2 Geometric imperfections C5. and plasticization of compressed concrete.1 in EN199211. 5.1. for increasing actions.linear elastic behaviour that assumes. see 5. independent of the importance of 2nd order effects. the solid line represents a member in one storey. The design procedures for linear analysis are given in [5. uncracked cross sections and perfect elasticity. the tolerance continues to decrease below the minimum value 1/300 when the number of storeys exceeds 8. Figure 5.non linear behaviour. the black line represents basic imperfection T0 according to expression 5. . for analysis. cracking. The EN rules are not repeated here. is to avoid the accumulation of discontinuities. EC3 and EC4. derived from both the linear and nonlinear analysis. Comparison between imperfection and tolerances. An upper limit 1/200 for the basic value has also been added. A good reason for having one value. the structure is represented by compressed and tensioned elements (strut and tie model).1 Symbols 5.1 General Structural analysis is the process of determination of the effects of actions (forces. For the total inclination of a column or wall in a structure.5EC2].linear elastic behaviour with limited redistribution [5.
5. a beam is a linear element. but also for isolated members. 6. The definition of the first order eccentricity in 4.8. This is important for e.the depth of the flange is at least 1/10 of the clear distance between ribs or 50 mm.8. which gives a fixed second order moment independent of the first order moment.3 Idealisation of the structure 5. ribbed or waffle slabs need not be treated as discrete elements for the purposes of analysis. formulations in 2.3 describe the imperfection rather as a first order effect. Thus. This may be assumed provided that (Fig. however.3 Structural models C5.3. Page 52 Table of contents . “outside” the second order analysis.6 and 5.it is the central part of a sensibly rectangular slab supported on four edges with a ratio of the longer to shorter span greater than 2. where an inclination has no effect on the column itself. whichever is the greater . Moreover: a slab subjected to dominantly uniformly distributed loads may be considered to be oneway spanning if either (Fig. The eccentricity can then represent either an uncertainty in the position of the axial load.2): . 4. 7. which indicates that it is not regarded as a first order effect.1. or an initial deflection (outofstraightness). or .5. since there is a link between imperfections and tolerances. an eccentricity or an initial deflection.3. like in the ENV. a pinended column.1. see 5.2.8.1 Classification of structural elements For buildings. or added separately.2): . In other methods. Otherwise it should be considered as a deep beam. On the other hand. the inclination can be transformed to either an equivalent eccentricity (or initial deflection) ei or a horizontal force Hi.2 further underlines this. see the definition in 5.the depth of the rib below the flange does not exceed 4 times its width .3 Equivalent eccentricity or horizontal force For isolated members. 4. In 4.Guide to EC2 Section 5 mean that the imperfection is on the “unsafe side”.1 in EN chapter 5.the rib spacing does not exceed 1500 mm .2.4 Dealing with first order effects Imperfections can be treated as first order effects.4 P(1) in the ENV.g. whereas Dim represents the average inclination of all vertical members contributing to a certain effect. for which the span is not less than 3 times the overall section depth.1 Structural models for overall analysis C5.6.5. With a method like the “curvature method” (“model column” method in ENV). the imperfection is associated with “uncertainties in the prediction of second order effects”. provided that the flange or structural topping and transverse ribs have sufficient torsional stiffness. It is essential to have a clear definition in this respect for the overall analysis of structures.it possesses two free (unsupported) and sensibly parallel edges.transverse ribs are provided at a clear spacing not exceeding 10 times the overall depth of the slab.3. and the horizontal force should give the same bending moment as the eccentricity (see figure 5.3. when other methods than the curvature method are used. there is no difference between the two approaches. as a convention. the second order effects depend on the first order effects.5. This is logical. or be added as separate safety elements without any physical meaning.7. The tolerance has to be checked for individual columns and walls. 5. This corresponds to a physical interpretation of the imperfection as a deviation in the form of an inclination. The equivalent eccentricity is linked to the effective length. a slab is a bidimensional member for which the minimum panel dimension is not less than 5 times the overall slab thickness. the ENV is ambiguous and unclear in this respect.2): Equivalent eccentricity: ei = Cantilever: ei = Pinended: ei = i i i l0/2 (l0 is the effective length) Hi = i i l l/2 Mi = N ei = N i l = Hi l Mi = N ei = N i N N l/2 = Hi l/4 Hi =2 5. and then it does make a difference whether imperfections are treated as first order effects. the following provisions apply: 5. In the EN it is stated once and for all that imperfections are to be treated as first order effects.
which may be analysed as solid slabs provided that the in situ transverse ribs are provided with continuous reinforcement through the precast longitudinal ribs and at a spacing according to Table [10.1) .3(11)EC2] in relation to prefabricated slabs without topping.3. This exception applies for slabs with clay blocks only. Parameters to determine effective flange width [fig. Figure 5. and in particular the assumption that plane sections remain plane.2. Hence different sections have different effective width. An exception to this rule is given at [10.3.4.3.3 – EC2] Page 53 Table of contents . 5.2 Geometric data C5. bw. for continuous beams. Note that the flange depth is not relevant. on the type of loading Fig. and that the distance between points of zero moment depends.1EC2] as a function of the cross section geometry (b. Point (4) makes clear that a constant width may be assumed over the whole span. The value applicable to the span section should be adopted. A column is a member for which the section depth does not exceed 4 times its width and the height is at least 3 times the section depth. the actual distribution of stresses is usually replaced by a conventional block.1 Effective width of flanges (all limit states) An exact calculation shows that the actual distribution of compressive stresses has a higher concentration in the part of flange which is close to the rib. In order to simplify calculations. 5.Guide to EC2 Section 5 s hf hw st Figure. Definition of lo for calculation of the effective flange width [Fig. the width of flange that effectively works with the rib in absorbing the compressive force (effective width) should be assessed.3.9. 5. 5.2 – EC2]. Figure 5. Otherwise it should be considered as a wall. width of ribs) and of the distance lo between points of zero moment. Effective width is defined at [5.1 Effective width of flanges of T beams (valid for all limit states) If a T beam with a relatively wide flange is subjected to bending moment. 5.EC2]. but is higher on both sides of the rib.3. This allows the application of the usual design rules.2.2 Geometric data 5.2.3.2. and a progressive reduction in the further parts. distance between adjacent ribs. This implies that the conservation of plane sections is not respected and that the neutral axis is not rectilinear.2EC2 is an example of a continuous beam (subjected to a uniform load distribution) where the lo distance for spans and for parts on supports is identified.Geometric parameters for slabs d 1500 mm t sn/10 or 50 mm d 4 bm d 10 h0 The minimum flange thickness of 50 mm may be reduced to 40 mm where permanent blocks are incorporated between the ribs. C5. It does not apply for expanded polystyrene blocks. even if it is expressly cited in (1)P. extended to the effective width.
in relation with continuous beams and slabs with ratio between adjacent spans in the range [0. For instance. Regardless of the method of analysis used. for which xu/d 0. without which the effects of impressed deformations would be devastating and quantitatively incorrect. respectively of average and high ductility.sup t / 8 where FEd.70 [5. shear). For the serviceability limit state (SLS) a gradual evolution of cracking should be considered. for concrete up to C50 and reinforcing steel of type B and C. necessary to ensure equilibrium.5 . These rotations transfer to other zones the effect of further load increase.sup ¸ = Ed. smaller than the moment M resulting from elastic linear design.5 Linear elastic analysis with limited redistribution [5. xlinear stressstrain relationships and xmean value of the modulus of elasticity. In the previous explanation the expression For the determination of load effects’ was used.sup 8 © t ¹ 2 4 5.45. calculated on the basis of a span equal to the centretocentre distance between supports. Two important points must be noted: Where a beam or slab is monolithic with its supports. the critical design moment at the support may be taken as that at the face of the support.Guide to EC2 Section 5 5. the fact that no limits were set to (xu/d) for the application of the linear analysis method at the ultimate limit states. It must be remembered that increasing values of xu/d the model uncertainty also increases and higher safety factors should be assumed for precaution. whereas [5.4(3)EC2] admits different assumptions.3.4(3)EC2]. stresses are proportional to loads and therefore the superposition principle applies. For thermal deformation. thus allowing to take.10a – EC2] (4. t 0. may be assumed in accordance with [5. the expression is: t 0.44 + 1. including thermal deformation. a reduced bending moment M.sup is the design support reaction t is the breadth of the support. therefore of all actions.3.2 Effective span of beams and slabs in buildings C5.EC2] At ultimate limit state plastic rotations occur at the most stressed sections. settlement and shrinkage for which [5. where a beam or slab is continuous over a support which may be considered to provide no restraint to rotation.8) where: G is the ratio of the redistributed moment to the elastic bending moment xu is the depth of the neutral axis at the ultimate limit state after redistribution d is the effective depth of the section Page 54 Table of contents . neglecting tension stiffening but including the effects of creep. does not mean that any value of (xu/d) may be used in design: it's opportune to observe a limit consistent with the method of linear elastic analysis with limited redistribution.5 – 2] expressions are given for the redistribution factor in function of the concrete class. Despite being named “Linear analysis with limited redistribution”. this is a design method. linear analysis may be used assuming: xuncracked cross sections.2. may be reduced by: MEd = FEd.4 Linear elastic analysis For the determination of load effects.5(4)EC2]. are considered. the type of steel and the xu/d ratio after redistribution. provided that in the other parts of the structure the corresponding variations of load effects (viz. Comments. for the design of reinforcement.25 (xu/d) .2.4(3)EC2] admits “For the determination of action effects”. settlement and shrinkage effects at the ultimate limit state (ULS).sup over the breadth of the support t §F · t t F MEd = ¨ Ed. the design support moment. With these assumptions.5 Linear analysis with limited redistribution C5. 5. The design moment and reaction transferred to the supporting element should be taken as the greater of the elastic or redistributed values. a reduced stiffness corresponding to the cracked sections.4 Linear elastic analysis C5. This formula derives by assuming an uniform distribution of the design support reaction Fed. That moment should not be less than 65% that of the full fixed end moment. taking into account the different types of support.2 Effective span of beams and slabs in buildings This paragraph defines the effective span length mainly for member analysis. In clause [5.
6. which states: “whichever load Q. at any section (xu /d) d 0. and integrating the curvatures along the axis of the elements. Other applications are the management of shear by the method of varying and the analysis of slabs by the equivalent frame analysis method [Annex I – EC2]. frames and slabs (in this last case with the theory of yield lines. by checking the required rotations against those allowed in accordance with [Fig.6. the rotation capacity must be verified. Compatibility conditions are normally expressed by assigning to each section its moment – curvature law. 5. the calculations must be developed for each load condition: for each one it is conventionally assumed that the ultimate limit state is reached through a single proportional increase of the applied load. For beams.7 Nonlinear analysis C5.208 =1 per (xu/d) = 0.5(1) expression (3.25 for concrete strength classes d C50/60 (xu /d) d 0. is higher or equal to the ultimate load Qu". is lower or equal to the ultimate load Qu". 5. the structure at ultimate limit states becomes a mechanism of rigid elements connected by yield hinges.7(1) Fig. 5. of the elements and of the structure (cracking.6. 5. 3.2. for steel 3.4 – EC2]. Deformations due to shear are generally neglected. trilinear idealizations of the moment / rotations Page 55 Table of contents . to which corresponds a kinematically admissible mechanism of collapse.1 Static method It is based on the static theorem of the theory of plasticity.45 and It must be considered that a redistribution carried out in observance of the ductility rules only ensures equilibrium at the ultimate limit state.70 per (xu/d) = 0. because it is a process of analysis.Guide to EC2 Section 5 The limits of this formula are: = 0. For the design of columns the elastic moments from frame action should be used without any redistribution. In the case of elements mainly subjected to bending. 3.6 Plastic analysis Plastic analysis should be based either on the lower bound (static) method or on the upper bound (kinematic) method for the check at ULS only. second order effects). clause [5. Resulting stresses are not proportional to the applied actions.14) and Fig. The process is developed by computeraided calculations. An important application of this method is the strutandtie scheme [5.7 Nonlinear analysis Nonlinear analysis is a procedure for calculation of action effects. Very high redistributions.1. Eurocode 2. based on idealisations of the nonlinear behaviour of materials [nonlinear constitutive laws: for concrete cf. Serviceability limit states requirements should be checked by specific verifications. to which a statically admissible tension field corresponds. The expression “statically admissible” indicates a field that meets both the conditions of equilibrium and the boundary condition without exceeding the plastic resistance.15 for concrete strength classes t C55/67 ii) reinforcing steel is either Class B or C iii) the ratio of the moments at intermediate supports to the moments in the span shall be between 0.5 and 2. 5. 3. The method is applied for continuous beams. It requires that the section geometry and reinforcement are defined. which states: “every load Q.6NEC2].2.6 Plastic analysis C5. by verifying equilibrium and compatibility at every load increase. It should be remembered that the plastic analysis methods shall only be used for checking ultimate limit states. Inelastic rotations are generally concentrated in the critical sections. those in relation with axial load are taken into account only in case have significant influence on the solution.2)EC2] states that the formation of plastic hinges is guaranteed provided that the following are fulfilled: i) the area of tensile reinforcement is limited such that.8]. very often must be lowered in order to meet the requirements of serviceability limit states.6. As the superposition principle does not apply because of the nonlinearity.2 Kinematic method In this method. which may be of advantage at the ultimate limit states. suitable for the nature of the structure and for the ultimate limit state under consideration. The method is based on the kinematic theorem. Specific verifications are needed for the serviceability limit states. If not all the conditions above are fulfilled.
Following the evolution of response to actions. In the text. where As is the cross section of the tensioned reinforcement. it is possible to verify the conditions for the serviceability limit states and for the ultimate limit state.3. First order effects First order effects are defined to include the effect of imperfections. this is the reason for using the word “nominal”. It can be defined as the difference between the ultimate moment resistance and the first order moment. The words in themselves are misleading.8.1 Definitions C5. third state (plastic): a third linear line can be idealized from the steel yielding clause to the point of failure moment. x the depth of the neutral axis. interpreted as physical deviations in the form of inclinations or eccentricities. then the nominal second order moment is greater than the true one.8 Analysis of second order effects with axial load 5. without load eccentricities or transverse loading.8. The ENV is ambiguous in this respect. The braced ones.8. hypothetical buckling of an initially straight member or structure. do not need to resist such forces. the others are braced. z the lever arm. all together. eccentricities and/or transverse loads. see 6. The rigidity can be increased by taking into account the contribution of concrete in tension between cracks (“tension stiffening”).1.8 are listed in 5. Sway – nonsway and global second order effects The terms sway – nonsway have been omitted in the final draft.5. have the necessary stiffness and resistance to develop stabilization forces.1.6. it ends when the tensional strength of concrete is reached (cracking moment) second state (cracked): from the cracking moment to the moment corresponding to steel yielding.Guide to EC2 Section 5 law of each critical section can be adopted as in Fig. This is also a reason why the word “buckling” is avoided in the title of 5. Some comments are given below. Bracing units/systems should be designed so that they. 4. Figure 5. Definitions Definitions specific to chapter 5. since all structures are more or less “sway”. buckling is mentioned only when a nominal buckling load is used as a parameter in certain calculation methods. with a value that can be deducted from the diagram in Fig 5. If the ultimate load is governed by instability before reaching the cross section resistance. Trilinear Moment – Rotation relation for element mainly subjected to bending 5. see also clause C5. but with caution in case of load cycles .2.6N of Eurocode 2 in function of the relative depth of the neutral axis. after many comments for or against.8 Second order effects with axial load C5. due to the presence of imperfections. Buckling The word buckling has been reserved for the “pure”. a Page 56 Table of contents . Braced – bracing The distinction braced – bracing is simple: units or systems that are assumed to contribute to the stabilization of the structure are bracing elements.8. representing the three following states: first state (elastic and linear): characterized by EI rigidity of the entirely reacting sections. to obtain a total moment used for design of cross sections to their ultimate moment resistance. The line corresponds to a pl plastic rotation at the critical section. moment increases are related to the curvature increases on the basis of rigidity Es As z(dx). by definition. Nominal second order moment The nominal second order moment is used in certain simplified methods. It is pointed out in a note that pure buckling is not a relevant limit state in real structures.
but without using the “sway” . These terms are now replaced by unbraced – braced. Basic criteria for neglecting second order effects Two basic criteria for ignoring second order effects have been discussed during the conversion process. The axial force is governed by vertical loads.3 Simplified criteria for second order effects 5. which are more general than those in ENVA.2. From this point of view.8.8.“nonsway” terminology. (See chapter 6 in this report for more details about interaction curves and the general method.1 (5). calculated according to the general method in 5.8.1 Slenderness criterion for isolated members C5. Two different ways of defining the basic 10%criterion for ignoring second order effects. and in the ENV.8.) 5.8.2. and in some cases misleading.3. The details are given in clause 3. A stiffness criterion like the one in ENVA. Figure 5. 2) 10 % reduction of the load capacity. Simplified criteria for ignoring 2nd order effects C5. This led to the present rules in 5. without the need for calculating them.1 General The load bearing capacity of a member in compression for low slenderness ratios is illustrated in figure 31 by means of interaction curves. (The interaction diagram was calculated for rectangular cross section 400 x 600 mm.8.6. see text above.3.3 and Informative Annex H. The classification of structures from this point of view remains in the EN.6.8.3.8.3.5.6 illustrates the consequences of these two criteria in an interaction diagram for axial force and bending moment.2 was avoided in earlier drafts of the EN.3. The basic criterion is further discussed in chapter 3 in connection with slenderness limits. during the conversion process there were many requests to include some simple criterion for evaluating the significance of global second order effects. namely: 1) 10 % increase of the corresponding first order effect. In a column or a structure it is the bending moment that is influenced by second order effects. Most design methods are based on calculating a bending moment. Figure 5. The second one has been claimed by some to be the “true”. including a second order moment if it is significant. edge distance of reinforcement 60 mm.2 General C5. hidden criterion behind the ENVrules. concrete C35.1 Slenderness limit for isolated members 5.1.3.Guide to EC2 Section 5 structure that would be classified as “sway” could be just as stiff as one classified as “nonsway”.2 (6). However. = 0. 5. The first criterion is the one stated in 5.3.1 (total mechanical reinforcement ratio). criterion 1 is the most logical and natural one.3.8.8. In the ENV the concept of sway – nonsway was linked to the criterion for neglecting global second order effects in structures.3. Their effects on the slenderness limit are discussed in chapter 3. assuming a constant eccentricity of the axial force.) Page 57 Table of contents . and is not significantly affected by second order effects. since it was considered as too crude. 4.
Criterion 1 will be more severe for high axial loads.1 the criterion was then identical to expression (4. Effective creep ratio Mef = 0. M01/M02 = 1) Depending on which of the two basic criteria is chosen. Therefore.2 A = 1 / (1+0.2 (6). In earlier drafts.3. see chapter 2. Mef = 0. Figure 5. see figure 5.8. a new model was developed. In a comment to the “final draft” it was pointed out that a limit independent of the axial force could be much on the unsafe side in certain cases (see next chapter).Guide to EC2 Section 5 Figure 5. calculated for a pinended column with cross section as for figure 21 and subjected to a constant first order moment.1: (1) As an alternative to 5. including the October 2001 “final draft”. 10 % reduction of load capacity for a constant eccentricity (Z = 0.3.8 Effect of normal force (here n = N/fcdAc) on slenderness limit. Equal first order end moments M01 = M02 = M0. will allow very high slenderness ratios for high axial loads. for Z = 0. when there is little room for bending moments. depending on the interpretation of the 10 % criterion: 1.62) in the ENV. on the other hand. one can find the slenderness ratios for which the basic 10%criterion is fulfilled. The following is quoted from 5. Criterion 2.3.7 and 5. By combining figures 5.13) Table of contents . a slenderness limit independent of the axial force was chosen as a compromise between the two basic criteria. Interaction curves for low slenderness ratios.2Mef) B = (1+ 2Z) / n Page 58 (5.8. 10 % increase of bending moments for a given normal force 2.8.6.7. increasing the axial force will either decrease (1) or increase (2) the slenderness limit as shown in figure 32. The following may be used: Olim = 20 A B C where: O slenderness ratio as defined in 5. second order effects may be ignored if the slenderness O is below a certain value Olim.8.
together with a general treatment of the slenderness limit. Comments on earlier versions of prEN 199211.for braced members with first order moments only or predominantly due to imperfections or transverse loading .7). Most people also considered it impractical and unnecessary to include creep. comments and examples were presented by Prof. was used in draft 1. Page 59 Table of contents . moment ratio M01.rm Mef effective creep ratio.1.8.10 shows examples of the effect of a rather moderate effective creep ratio.3. 10 % increase of bending moments due to second order effects 2. if Mef is not known. C = 0. C < 1. 10 % reduction of the load capacity for a given eccentricity It was then demonstrated that the effect of the normal force on the slenderness limit was different depending on which alternative was used.7): .3 Background to the new proposal After receiving the abovementioned comments and examples from Prof Hellesland. creep and moment ratio (different end moments). independent of the normal force. Figure 5.Guide to EC2 Section 5 C = 1.8. In the “final” draft October 2001 an interpolation was introduced to avoid discontinuity (expression (5. otherwise negative (i. a column bent in double curvature (end moments of different directions.0 (i. figure 5.3.e. relative normal force rm = M01/M02. The reinforcement ratio was not included then. In the following cases. and therefore the ENV criterion (4. e. Mef =1. 5. see 5.for unbraced members in general The background to the new criteria is presented in the following. C > 1.e. A = 0. An addition was made in draft 2.e. rm should be taken positive (i. a systematic investigation of the slenderness limit was made. M 02 t M 01 (2) If the end moments M01 and M02 give tension on the same side. rm should be taken as 1. combined with a high effect of creep and a moderate or high normal force. mechanical reinforcement ratio. there was a widespread opinion that creep would have little effect at these low slenderness ratios. M02 first order end moments.9. normal force. However. there was no agreement as to which alternative to base the slenderness limit on.7 may be used Z = Asfyd / (Acfcd). allowing the constant 25 to be increased to 35 if the reinforcement ratio Z is at least 0.2 / n may be used As total area of longitudinal reinforcement n = NEd / (Acfcd). There was also a lot of discussion about the interpretation of the basic 10% criterion for ignoring second order effects.13) might be extremely unsafe in some cases.2 History of the slenderness limit in prEN 1992 At an early stage of the conversion process (spring 1999). are given in [8]. see figure 5. B = 1.7 .g.13)). with the two main alternatives (see chapter 2): 1. J.9).1.5. since it was considered impractical. In November 2001. shortly after the draft had been distributed. a slenderness limit was proposed in which the effective creep ratio Mef and the relative normal force n were included as parameters. if Z is not known.7).62). Column bent in double curvature 5. Hellesland.4. Figure 5. with focus on the effects of reinforcement.8.8. showing that (5.
a default value based on a low value of has also been given. since the reinforcement is normally not known when the slenderness criterion is checked. and that is true for both 10%criteria.8 and the disagreement concerning the 10%criterion.) Table 5. the slenderness limit would have to be either very conservative for high to moderate values of Z. Z = 0.0 20 10 0 45 30 1.9 70 50 For n = 1.11.0 11 20 0 27 43 37 0.9 74 107 1. (Solid lines = 10%criterion alt. % Mef = 0 Mef = 1 1.11 and table 5. Page 510 Table of contents . Table 5. having in mind that Mef = 1 represents a rather moderate effect of creep. However.1 Z = 0.10. Values of slenderness limit taken from figure 5.11.10. Olim Average reduction n M01/M02 due to creep. or as a starting value in an iterative process.1.2. The only exception is when criterion 2 is applied to columns with equal end moments.1 to Z = 0. see figure 5. With a higher value of Mef the reduction is more severe (most of the comparisons were made with Mef = 0 and 2 respectively.4 0.1 shows some values taken from the figures.2. see Appendix 1). ( Mef = 2) Olim Average increase from M01/M02 Z = 0. as can be seen from figures 5. Slenderness limit as a function of the relative normal force and the moment distribution.0 17 27 0 47 70 16 0.3. These proposals were rejected until and including the October 2001 draft.5 0 70 60 16 0. referring to figure 5.11. Without Z as a parameter. Values of slenderness limit taken from figure 5.5 % Z = 0. Effect of reinforcement ratio on slenderness limit (only criterion 1 is shown here). alternative 1 (Z = 0. Concrete C40. the average reduction due to creep is considerable. or on the unsafe side for low values. with different end moments there is a strong reduction of the slenderness limit with increasing normal force.3.5 1. Table 5.Guide to EC2 Section 5 Curves according to both basic 10% criteria are shown.8 There have been national comments proposing to include the effect of n in the slenderness limit. dashed = alt. The effect of Z is considerable. Figure 5. This can be used for a conservative estimation.10.0. 2.0 37 0. However.0 30 25 0.9 41 66 n 0.9 110 90 1. Figure 5. as can be seen from both figure 5. 1.10 and 5.
0.9 110 90 87 1.8.3).4 0 70 60 60 0.4.9 74 107 1. Slenderness limit in draft Oct. but it severely overestimates the slenderness limit for a high normal force combined with a high creep effect. Slenderness limit according to new model compared with table 5.13): Olim = 25 (Z + 0.9 41 66 Draft October 2001 gives a fairly correct influence for Z.4.Guide to EC2 Section 5 5.3) Olim n M01/M02 10% criterion (alt.4 (here the values are based on Mef = 2).1 to Z = 0. Table 5.0 20 10 30 0 45 30 60 0.2 respectively.5 and Mef = 0 the values according to draft October 2001 are reasonable (table 5. values from tables 5. Slenderness limit according to new model compared with table 31 (Z = 0. however. Olim Average increase from n M01/M02 Z = 0. Table 5.8.5 1.5 and 5.3 and 5.5 % Z = 0.4 0 47 70 16 0.3.0 11 20 0 27 43 0.1 and 5.2 ( Mef = 2).3). particularly for Mef = 1 and most particularly for M01/M02 = 1.1) Draft Oct.3. 2001 Mef = 0 Mef = 1 1.1 (Z = 0.M01/M02) In both cases the 10 % criterion is according to alternative 1. 2001 compared to values from table 5.9 70 50 87 For n = 0. 4) is compared to the same data as in tables 5.0 17 27 0.8 0. they overestimate the slenderness limit.1 Z = 0. Slenderness limit in draft Oct. The omission of the effects of both normal force and creep are the main disadvantages of this model. see table 5.6 the new model (see p.3.5 Comparisons with new model In tables 5.5. For n = 1. where it gives a 3 times too high value. 2001 compared to values from table 5.8 37 0.0 30 25 30 0. 5.6.1. Table 5.1.1and 5.9) (2 .2 ( Mef = 2) Page 511 Table of contents . Table 5.2 are compared with values according to the slenderness criterion in the October 2001 draft (expression 5.4 Comparisons with previous model for slenderness limit In tables 5.
It has been claimed that the conservativeness was deliberate.2 Slenderness and effective length of isolated members C5. in order to cover certain unfavourable nonlinear effects.2 (4). draft 1 Figure 5. 5. The importance of considering creep in the slenderness limit is further substantiated in 4.1. e.3. flexible foundations etc. Effective length according to accurate and simplified calculations. by also including the concrete grade. without including some hidden allowance for possible unfavourable effects.15) and (5. 1). the new model gives good agreement with the 10%criterion (alt.12b. based on the definition of effective length in 5.2 (5) and in 5. The expressions in draft 1 were taken from UK proposals.8. However. Page 512 Table of contents .12a and b show comparison between an accurate numerical calculation of the effective length and estimations according to draft 1 (a) and final draft (b) respectively.7.8.13 and 5. A complete verification of the new model is given in Appendix 1.4) and in both cases the values are conservative compared to 10%criterion alt.2 Effective length New expressions (5. also for Z = 0.g. However.12a. The present kfactors express the relative flexibility of the restraint according to the definitions in figures 5. included in comments on the ENV and on earlier EN drafts. Derived to give accurate estimation. Concrete grade is indirectly taken into account in the n value.3.8. It was found that they are very conservative in some cases. were introduced in the second draft. 2. as in the case of unbraced members with the previous expressions. table 5. Figure 5. and the present models are aimed at giving an accurate estimation according to this. final version The present kfactors are defined differently compared to the corresponding factors in draft 1 and ENV.16) for the effective length of isolated members in frames.1. There is a slight overestimation of the slenderness limit for n = 1.8.8.3.8.1 and n = 0. December 1999. Such effects are instead explicitly addressed in 5.8.6. but the present model is considered to be good enough.0 and Mef = 1.3.6 Discussion On the whole.5. Figure 5. table 5. giving up to 40 % overestimation of the effective length for braced members and on the unsafe side in other cases. The new expressions also avoid unsafe estimations. the overestimations are small compared to the old model (see 3. and are called k1 and k2 to avoid confusion with the previous factors ka and kb. they replace figure 4. and the main parameters are taken well into account. giving up to 20 % underestimation of the effective length for unbraced members.Guide to EC2 Section 5 5.23) in draft 1 of the prEN.3.14. A more sophisticated model could of course give even better agreement. Effective length according to accurate and simplified calculations. the effective length is by definition based on linear behaviour.22) and (5. They are applicable to different types of flexible moment restraint.27 in the ENV as well as expressions (5.1. such as beams with different boundary conditions.3.
If both columns connected to the node reach their respective buckling load at the same time (under proportional increase of loads)..15. or as contributing to the restraint. This will depend on the magnitude of the axial force in the adjacent column.2 (4) addresses the question whether an adjacent column (in a storey above or below) in a node should be considered as using the same restraint as the column considered. when the adjacent column has a relatively low axial load.D . i. they will both have to share the restraint provided by other connected members (beams). example of definition. Comparison between different definitions of kfactors. it will contribute to the restraint.13. Flexibility of restraint. In the opposite case. examples. it can be included among the members which resist the moment M..8.e. A reasonable model for the transition between the two limiting cases gives by the following: k= § EI EI · T ¨ D a + c ¸ M1 + M 2 + .+ 1 .14. Figure 5.Guide to EC2 Section 5 Figure 5.3. 5. and k should then be defined as k= T ª EIa EIc º + « » M ¬ la lc ¼ (31) Here subscripts a and c refer respectively to the adjacent column and to the one considered see Figure 5.
15 Ma restraining moment in the adjacent column.15. 2…. calculated without taking into account the axial force Na D = Na/NBa Na axial force on the adjacent column NBa buckling load of the adjacent column (can be estimated approximately. taking into account only the horizontal members adjacent to its nodes) Page 513 Table of contents . M a © l a lc ¹ (32) where M1. see Figure 5. e. M2… restraining moments in members 1. see Figure 5.g.
Since the limitations of the applicability are not stated. In practice.3 (3): M Ed  M0Ed d 1. rigid moment restraint The basic criterion “second order effects 10% of first order effects” gives.3.8.Guide to EC2 Section 5 Figure 5. there is a also risk that it is used outside the scope. and further extension to the “full version” are given in Annex H.BB 0. 5.3. No significant global shear deformations (e. together with the simplified magnification factor for bending moment in 5. However. two alternative proposals were presented to CEN TC250/SC2 (Berlin. • “Full version”: formulated in a more general and transparent way. no significant openings in shear walls) These conditions are not fulfilled for e. expression (H.3 Global second order effects in structures 5.1 § 0.2 According to the ENV. After many requests to include something similar.3.2. Detailed information is given only for regular cases. A.2 has a very limited field of application.1 M0Ed 1 . cf.3 Global second order effects in buildings C5.1 Background a) Model in ENV. A simple extension is given to cover the effect of global shear deformations.1): FV.8. global second order effects may be ignored.8. No significant rotation at the base (rigid restraint / stiff foundation) 2.7. In the final draft a somewhat extended “miniversion” is given in the main text. a bracing system including frames.1 / 1.BB (35) Here Page 514 FV. and the criterion is improved to be less conservative and to take into account cracking in a simple way.3. giving unsafe results. a “miniversion” and a “full version”: • “Miniversion”: same scope as in ENVA. The criterion also explicitly requires that bracing members are uncracked.3.3.Ed total vertical load Table of contents . For the above reasons.FV.3. Illustration of node with adjacent members. the conditions and restrictions are clearly stated. the ENV criterion was not included in earlier drafts of the EN. and no information is given for the cracked stage. 5. due to high lateral and low vertical loadings (most of the vertical load is often carried by the braced members). and a criterion given in a similar closed form. b) New proposals Due to the above shortcomings.Ed FV.Ed FV.g. but the formulation opens for general cases. if Where L FV EcmIc D ns (33) L = FV Ecm Ic d D total height of building (htot in the ENV) total vertical load (FV in the ENV) sum of bending stiffnesses of bracing members 0.1 FV.BB (34) This gives the following criterion for the vertical load.6 (in the ENV.3.3. May 2000).2 No significant shear deformations. ENV A.8.15.1ns 0.2 + 0. no particular symbol is used for this parameter) number of storeys (n in the ENV) This criterion is valid only on certain conditions (not stated in the ENV): 1. bracing members are often more or less cracked in ULS. shear walls with large openings and/or flexible foundations. and the structure may be considered as “nonsway”.8.g.
calculated for a secant stiffness representing the relevant ULS conditions (including lateral loading). Constant stiffness and equal increment of vertical load per storey. and with a normal force there is a more or less smooth transition between the two stages.8 ns ns +1.16 (the buckling load has been calculated numerically by Vianello’s method. In a structure. The coefficient 0. usually associated with high vertical load. For constant stiffness. a higher stiffness value for uncracked section is not given in 5. Then there effects not only of cracking. to account for cracking in a simplified way 6EI is based on 0. and for uncracked section 0.4 Ecd Ic ns E I E I cd c = cd 2 c = 0. equal load increment per storey and rigid moment restraint at the base.BB nominal buckling load for global bending (no shear deformations) M0Ed first order moment MEd design moment The global buckling load for bending can be written FV. It is not a load for which “pure” buckling (without eccentricities or lateral loading) would occur.16 can be approximated by [0  7.4 will be low.16. which gives less overall effect of cracking. However. where all the vertical load considered acts on the member itself.1 [0 0.6) 1 The ratio 0. Figure 5. there is little effect of creep. particularly in cases where the section is uncracked.6 where ns = number of storeys Combining expressions (34) to (36) gives FV d 0. which means that there is less effect of compression nonlinearity on the bracing units. the effective creep ratio according to 5.26).8. [0 will depend on the number of storeys and (to some extent) on the distribution of vertical load between braced and bracing members according to Figure 5. Global buckling due to bending and coefficient for buckling load.BB is a nominal buckling load.4 L total height Note FV.312 ns +1.5 between the stiffnesses for cracked and uncracked sections is of course a rough simplification. but also of nonlinearity in compression are considered. The reason is that for global second order effects in structures. Page 515 Table of contents .4 and = 0. and [0 has then been evaluated according to expression (36)).1. A further difference is that the bending moment normally has a more favourable distribution in a bracing unit than in isolated members. on the other hand.26) is valid for isolated members.8 0. For the same reason.312 ns /(ns + 1. and consequently.4 (or 0. Creep is not included in the criterion for neglecting second order effects in structures (as it is for isolated members).3/(1+Mef) in expression (5. the dominating first order effect is wind. since this is about cases where second order effects are more or less negligible. most of the vertical load is normally on the braced units.8.4 EcdIc. The ratio should depend on the reinforcement and the normal force.6 L2 L2 L (38) where 0. In this circumstance. The coefficient [0 according to the upper curve in Figure 5. The coefficient [0 in expression (36) depends on various parameters.312 = 7. These circumstances together justify the use of 0.3/(1+ ef). in which case a particular value for uncracked section (0.8) for estimating the stiffness (see H.8 may be used instead of 0.1 0. The last effect can be strong.2. a simple rule is acceptable. Expression (5. distribution of vertical load etc.8) is justified1.Guide to EC2 Section 5 FV.2 (3)) can be compared to 0.BB = [0 ¦ EI L2 (36) where [ coefficient depending on number of storeys.4/0. 6EI total bending stiffness of bracing members.7.8 instead of 0.
3. Figure 5.2) in EN annex H.3 Effect of flexible moment restraint Flexible moment restraint at the base will reduce the buckling load. however.2) L EI T = rotation for bending moment M (compare figure 5.3. a direct reduction of the buckling load is more convenient. see figure 5.18) can be formulated in the same way (substituting Ecd with Ecm and explicitly including partial factor JcE = 1. In the global analysis of structures.19): The (hypothetical) buckling load for shear deformations only is: FV. The comparison shows that for uncracked section. the two models give rather similar results.13) The factor [1 is an approximation. curves for both cracked and uncracked sections are shown.8. the isolated effect of flexible end restraint can be accurately modelled by the factor [1  1 1 0.17 the two corresponding parameters /1.18. The ENV gives no values for cracked section.17. although the ENV is often much more conservative. The product [0[1 corresponds to [ in expression (H. Solid curves represent the “exact” solution. see figure H1 in the EN).18).2 on the right hand side): L = FV Ecm Ic d 1.18. Figure 5.Guide to EC2 Section 5 This is the background to expression (5.3. For the EN. This effect can be directly included in the [–factor for the buckling load. Page 516 Table of contents . which has been derived by calibration against accurate numerical calculation for different numbers of storeys. Effect of global shear deformations a) Shear deformations only (see Figure 5. (For isolated members this effect is normally taken into account by increasing the effective length. therefore there is no comparison for this case. Compare the ENV formulation (see 3. reasonably equal increment of the vertical load per storey and constant cross section.2 (EN) and D2 (ENV) are compared.1 above): L = FV Ecm Ic d D (310) Expression (5.7k (312) where k= TM (same definition as for isolated members. therefore no comparison can be made. see 3. Comparison between EN and ENV criteria 5. dashed curve the approximation according to expression (312) The effect of flexible moment restraint is not covered in the ENV. Effect of flexibility of end restraint for bracing units. since the effective length is not a practical parameter for bracing members and systems having varying axial load).2 (311) In Figure 5. For bracing units with two or more storeys.BS = S (or 6S for more than one bracing unit) (313) Here S is the shear stiffness (= shear force giving a shear angle = 1.
Ee = Ec/(1+M).4. CD . assuming linear elastic material behaviour: a) uncracked unreinforced cross section (5.BB FV .20. AB .additional load (QD .20 illustrates a hypothetical load history and the corresponding deformations.8.longterm load QL giving an elastic deformation 2.8. The basic criterion for neglecting second order effects is the same as before: FV 0. This corresponds to line AC in figure 5. taking into account bending and shear deformations. The total load is assumed to consist of one Longterm part QL (corresponding to the quasipermanent combination) and one additional shortterm part up to the Design load QD. on the other hand. it can be verified also for bracing units with vertically increasing axial load and significant global shear deformations (e. three examples are used to derive and illustrate the effective creep ratio ef.8. Hypothetical buckling due to global shear deformations only b) Combined bending and shear: The combined buckling load.2) c) cracked reinforced section (5.2 The total load history can then be divided into three parts: 1.BB FV.1 General The ENV stated that creep should be considered in connection with second order effects. Effective creep ratio 5.B (315) which leads to expression (H. With the effective creep ratio.2. the analysis can instead be made directly for the design load in one step. The “effective equivalent concrete modulus” would then be Eef = Ec/(1+ Mef) where Mef is the effective creep ratio.BB 1 = = 1 FV.4 Creep C5.20.4. By numerical calculations.8.3) Page 517 Table of contents .2.4.8.BB ¦S (314) Expression (314) can be derived analytically for simple cases like isolated members with constant normal force.6) in Annex H.2 Effect of creep in cross sections In the following.BS 1 + FV.19.20.8. shear walls with large openings).4. can be expressed as FVB  FV.Guide to EC2 Section 5 Figure 5. therefore no comparison can be made. BC . based on the so called “effective creep ratio”. but gave no information on how.2.4 The total deformation under design load can be calculated in a similar way if an effective creep ratio Mef is used. applied after a “long time”.BS 1 + FV. Figure 5. A general approach would be to first calculate creep deformations under longterm load. 5. line AD in figure 5.4.8. In the EN.g. The examples deal with bending moment and curvature in the following cases. practical models for taking into account creep are given.1 FV. This case is not covered in the ENV.4.1) b) uncracked reinforced section (5. then to analyse the structure for the additional load up to design load. 5.constant load QL giving a creep deformation based on full creep coefficient M 3.BB +1 FV.QL) giving an additional elastic deformation Figure 5. Illustration of load history and deformations The total deformation under longterm load can also be calculated directly using an equivalent Emodulus3 for the concrete.
The total curvature under a longterm bending moment ML is (cf.2.20): ML §1 · 1 + M . line AC in figure 5.8.4.Guide to EC2 Section 5 5.1 Uncracked unreinforced cross section This is the simplest case for demonstrating the idea behind the effective creep ratio.
the total curvature will be MD § 1 · MD ML MD § M L · MD §1 · ¨ r ¸ = E I ¨ r ¸ = E I M E I = E I ¨ 1 M M ¸ = E I 1 Mef . ¨ ¸ = © r ¹ L Ec Ic (41) The part caused by creep can be separated: ML §1 · ¨ ¸ =M E c Ic © r ¹C (42) Under design load with total bending moment MD. part of which is a longterm moment ML with a load history according to figure 5.20.
8. using an equivalent Emodulus for concrete (see second footnote in 5.2 Uncracked reinforced cross section The total curvature under longterm bending moment ML can be expressed in the following simplified way. with different notation): Mef = M ML MD (44) 5.19).2. © ¹D © ¹C c c c c c c c c © D ¹ c c (43) Thus.20): § 1· ¨ ¸ © r ¹L ML EI ML Ec Ic E s I s 1 M (45) The part of this curvature caused by creep can be separated: § 1· ¨ ¸ © r ¹C ML EI ML ML Ec E c Ic E s I s Ic E s I s 1 M (46) Introducing the following parameters: = (is/ic)2 D = (Es/Ec)(As/Ac) ic = radius of gyration of concrete area is = radius of gyration of reinforcement area curvatures can be expressed in the following way: ML 1+M §1 · ¨ ¸ = © r ¹L Ec Ic 1+ 1 + M .4. expression (5. the effective creep ratio is (cf.
ML §1 · ¨r ¸ =E I © ¹C c c § 1+M 1 ¨ ¨ 1 + 1 + M .
1 + © · ¸ ¸ ¹ (47) (48) Under design load and total bending moment MD the total curvature will be. including creep due to a longterm bending moment ML: MD 1 § 1 · MD 1 ML §1 · ¨ r ¸ = E I 1+ + ¨ r ¸ = E I 1+ + E I © ¹D © ¹C c c c c c c § 1+M 1 ¨ ¨ 1 + 1 + M .
see expression (47): 1 + Mef MD §1 · ¨ ¸ = © r ¹D Ec Ic 1 + 1 + Mef . 1 + © · ¸ ¸ ¹ (49) The same curvature expressed with the effective creep ratio would be.
after simplification: 1 + Mef 1 + 1 + M . (410) Combining expressions (49) and (410).
= ef M 1 + L 1+ MD § 1+M 1 ¨ ¨ 1 + 1 + M .
1 + © · ¸ ¸ ¹ (411) From this the effective creep ratio can be solved: Mef = 2 Subscripts 3 A 1 + DU .
see clause 4. However. 1 1 .4. Nonlinear effect will be dealt with later.3. since concrete stresses will decrease and reinforcement stresses increase with time. 4 Theoretically this is not fully correct. it is a reasonable approximation in most cases.ADU (412) L and D are used in this chapter for simplicity. Page 518 Table of contents .8. they correspond to Eqp and Ed in 5.
Guide to EC2 Section 5 where A= M 1 + L 1+ MD § 1+M 1 ¨ ¨ 1 + 1 + M .
21 shows the relationship between Mef and ML/MD for M = 3 and different values of for the uncracked reinforced cross section. 5.19) becomes more or less conservative.4. i. Effective creep ratio as a function of the ratio ML/MD for = 3. As the simplest case.19). > 0. consider a rectangular cross section with bending moment only and tensile reinforcement only. With reinforcement.e.2. uncracked rectangular cross section with symmetric reinforcement. The flexural stiffness in the cracked stage (ignoring any contributions from concrete in tension) can then be expressed as EI Es As d 2 1 [ . Figure 5. although it is a little more complicated. expression (5.3 Cracked reinforced cross section The cracked cross section can be treated analogously.8. edge distance t = 0.1h and D =Es/Ec = 6 The straight line for = 0 is identical with expressions (44) and (5. 1 + © · ¸ ¸ ¹ Figure 5.21.
1 [ 3 .
(413) where As = area of reinforcement d = effective depth [ = x/d x = depth of compression zone The relative depth of compression zone for a certain creep coefficient M can be obtained from § · 2 1 ¸ [M = 1 + M .
D ¨ 1 + ¨ 1 + M .
[ M 3 . introduce the symbol BM = 1 . D ¸ © ¹ (414) where D = (Es/Ec)(As/Ac) To simplify expressions.[M 1 .
1 .
.
(415) The total curvature under design load can then be written: MD ML MD §1 · B0 + BM . Values of Mef for which expression (416) is satisfied can be found by iteration (direct solution is not possible in this case). Figure 5.22 shows the result for M = 3.B0 = BMef ¨ ¸ = r ¹D Es As d 2 Es As d 2 Es As d 2 © (416) where B0 is parameter according to expression (415) for M = 0 and B ef is the same for M = Mef. Page 519 Table of contents .
5. Figure 5.8. After calculating point C. Therefore. As a simplification. b.y2. A slender column behaves in a nonlinear way. based on d = 0.Guide to EC2 Section 5 Figure 5.4. including the relevant first order moments or eccentricities for each step.3.23. Longterm load QL during time tt0. steps 1 and 2 can be combined into one.4 is illustrated in figure 5. curves for low and moderate ratios are also quite close.2. the additional load QD – QL is added.19) the higher the reinforcement ratio is.4 Conclusions concerning cross sections The idea behind the effective creep ratio in 5.9h and D = 6. with deformation starting from y2. due to both material and geometrical nonlinearity. Effective creep ratio as a function of ratio ML/MD for a cracked rectangular cross section with tensile reinforcement only.y1 as an initial deflection added to other first order effects. See line CD in figure 5. see 6. the total load QD is applied “from scratch”. Illustration of load history and deformations with nonlinear behaviour The load history can be divided into three steps: 1. immediate deformation y1. calculated for Mef = 0 2. will be used in twostep calculations in the following way.3 Effect of creep in slender columns 5.23. but with y0 = y2 . However. in a reinforced section the overall effect of creep on stiffness is reduced with increasing reinforcement. Page 520 Table of contents . additional deformation y3 .22. The simple linear relationship according to expression (5. Application of longterm load QL. This corresponds to line AC.8. and the calculation is then reduced to two steps.8.20 is outlined in figure 5. The last step can be calculated in two alternative ways: a. Alternative b. total deformation y2.4. A nonlinear behaviour similar to the linear one in figure 5.1 General The above conclusion concern only cross sections and are based on linear material behaviour. 5.23. See line ED in figure 5. Load increase up to design load QD. In this point the relevance of the effective creep ratio for slender columns are examined. the effect of deviations on the stiffness will not be as high as it may appear from the above figures.4.23.19) is always more or less conservative. using a stressstrain diagram with the strains multiplied by (1+M). The distribution of y0 along the column should in principle be the same as the distribution of y2 – y1. calculated for Mef = 0 A realistic calculation representing this load history should involve these three steps. Furthermore. For a pinended column.4. since creep only affects the concrete contribution to the stiffness.8. calculated for Mef = M 3. Basic creep coefficient M = 3 In this case the curves will approach the straight line according to expression (5. but deviations are generally small.20 and demonstrated in three examples. After calculating point C.
Guide to EC2 Section 5 however.e.125 1. A high slenderness ratio has been chosen. alternative a) will be the normal choice in practical design. All calculations below have been done with the general method.08h (same for longterm and design load.4.Concrete C40 .023 = 0. calculated with M = 3: 3. For the definition of ef there are two main options: a) based on first order moments M0L and M0D. The reason is that the second order moment is a nonlinear function of the axial load. which depends primarily on bending moment. i. see chapter 5. Mef = M ML / MD The relevant deformation parameter in second order analysis is curvature. and the ratio ML/MD will be lower if second order moments are included.111 (creep deformation) The load capacity calculated with e0+ y0 and with M = 0 is nRd = 0.0173 = 0. The load capacity under this total first order effect and no creep (M = 0) is calculated. would lead to the wrong conclusion here. therefore the effective creep ratio can be based on axial loads as well as moments. no absolute dimensions are used. y0/h = 0.1h .28 6 The load capacity with Mef = 1. total deformation.e.Mechanical reinforcement ratio Z = 0. y1/h = 0.Reinforcement S500 .235 b): nL = 0.235 = 1.15 (total reinforcement) . Therefore. Mef = M M0L / M0D b) based on total moments ML and MD. (for a general description and discussion of this method.0173 2.134 (M = 3) 3. using the effective creep ratio. y2/h = 0.0819 – 0.235 in twostep calculation. In the general case only moments should be used. A further simplification is a onestep calculation. This is easy to verify with a magnification factor based on linear material behaviour (see chapter 5.0819 y0/h = 0.100/0. Geometric assumption are: .023 (M = 0) 2. a sineshaped or parabolic distribution will be adopted as a simplification.8.8.2 Comparison between one.Basic creep coefficient M = 3 In the following. Alternative b) is the most realistic one.5 5.8. the moment increase due to second order effects will be greater under design load than under longterm load. a): nL = 0.Eccentricity e0 = 0. Alternative a. calculated with M= 0: y1/h = 0.100 Mef = M M0L / M0D = M NL e0 / (ND e0) = M nL / nD = 3 0. i. which depends on total moments etc. and is added to the constant first order eccentricity e0 given above.134 – 0.Edge distance of reinforcement 0.100 (longterm axial load) 1. in order to emphasize the effects considered. since creep deformations will mainly be governed by total moments. including 2nd order moments. which depends on effective creep ratio. n = N / Acfcd and m = M / hAcfcd.7). all axial loads and bending moments are expressed in relative terms.198 Cf.e. Therefore. this tendency will be even stronger in a nonlinear analysis.and twostep calculations An example will be used to compare the onestep calculation.189 These values are compared to the result of a onestep calculation. giving a fixed 2nd order moment. using an effective creep ratio Mef. Therefore. Therefore.0646 y0 is taken as an initial deflection with parabolic distribution. with the more realistic twostep calculation. creep deformation: y2/h = 0. iteration is inevitable since second order moments depend on stiffness. i.Slenderness l/h = 40 (O= 139) . Table of contents Page 521 . (line AD in the figure). 0.Rectangular cross section with reinforcement concentrated to opposite sides . In this particular example the first order moment is proportional to the axial load.28 is nRd = 0. With this alternative. no other first order effect) . immediate deformation.3. thus the result is 16% conservative 5A 6 "curvature method". a) nL = 0. using an effective creep ratio based on first order moments. the axial load should not be included in the definition of effective creep ratio.6). The result is nRd = 0. is always more or less on the safe side. however.
0162/0.3.0618) This is within 5 % of the twostep calculation (which gave nRd = 0.7.8.134) = 0. comparing the onestep and twosteps methods according to 4.183 This is within 3 % of the twostep calculation (nRd = 0.189) 5. thus the result is 12 % conservative These results are somewhat conservative. The basic parameters are the same as for the example in 4.0162 After iteration the following values are found: Total moment under design load mD = 0.Guide to EC2 Section 5 b) nL = 0.98 nRd = 0.1.3 Conclusions It is conservative to use an effective creep ratio based on first order moments. however.7 shows the results of calculations.3. Therefore.3.125 Mef = 3 0.0267 After iteration: mD = 0.0618 = 0. however.1.4.2. here M = 3 has been assumed Page 522 Table of contents . This is further discussed below. total moments are much more complicated to use. dealing with the slenderness limit in general.100 (0.786 is nRd = 0. Table 5. It is thus a complement to clause 3.125 Total moment under nL is mL = nL (e0 + y2) = 0. 0.189 in twostep calculation. see 3. It is also a complement to 4.08 + 0.4.125/0. as could be expected (the reason is explained above).151.4 The effect of creep on slenderness limit The effect of creep on slenderness limit will be further studied here.3.100 Total moment under nL is mL = nL (e0 + y2) = 0.786 Load capacity with Mef = 0. In practical design.189 = 1.0618 Effective creep ratio Mef = M mL / mD = 3 0.3. The effect of creep for columns with a low slenderness. the normal procedure will be to use first order moments. 5. since iteration will be necessary.166 Cf.2. since this clause deals with low slenderness ratios. except those for which different values are given.0819) = 0.0267/0. Mef = M mL / mD = 3 0.0531. nRd = 0. Table 5.08 + 0.8. Next is a onestep calculation with Mef based on total moments.125 (0. total moments will give more accurate results.2.1 and 4. based on a slenderness corresponding to the limit for which second order effects may be neglected with Mef = 0.224 (total moment for this load is mD = 0.0531 = 1. a) nL = 0. Symbols in table: e02 the greater of the two first order eccentricities e01 the lesser eccentricity l0/h slenderness corresponding to the limit for 10 % moment increase at M = 0 n relative normal force N/Acfcd nu0 load capacity for the current slenderness and M = 0 nL longterm load Mef effective creep ratio = M nL / n0. dealing with creep combined with a high slenderness.235) b) nL = 0.
35 = 0. there is a significant effect of creep. The load capacity for ef = 0. higher first order moments and higher amounts of reinforcement. which is allowable at these slenderness ratios. see discussion in the first paragraph above. In most cases a first order eccentricity e02/h = 0.19 = 1.74M ? The example in 4. e. the highest possible effective creep ratio is Mef = M M0L/M0D = M 1.28). A reasonable lower limit for the load safety factor could be 1. with the aim of having a moderate normal force. longterm load = 74 % of design load is the worst possible case for consideration of the effect of creep. and since second order moments are small at these low slenderness ratios. Even with the low normal force.0/1.19 This safety factor may be considered somewhat low. The load capacity for full creep. and it can be shown that this factor will be higher for lower values of slenderness. For e01/e02 = 0.) Conclusion concerning the effect of creep on the slenderness limit In these examples.141/0. In a more accurate calculation they should be integrated from 0 to M. In these cases. is found to be nRdL = 0.5 Safety under longterm load only The effective creep ratio is based on moments under quasipermanent load which. creep reduces the load capacity by 5 to 30% (average 15%). with increasing second order moment. The corresponding longterm load is nL = 0. and then the 2steps method becomes uncertain. it is conservative to use an effective creep ratio based on first order moments. In normal cases there is always some variable load. Thus.141.134. the current example is rather extreme. however. However. in the unfavourable direction. is an SLS load with no load factors (except \2 < 1 for variable loads). e02/h = 1. since the stresses are normally not very high under longterm load. This is higher than the current long term load.35 NL with Mef = 0.022.2 is used again.Guide to EC2 Section 5 nu1 load capacity including the effect of creep according to 1.118.159.9 (double curvature bending) the 1step method is generally slightly conservative compared to the 2steps method.134/0.22 is found to be nRd = 0. The percentage of longterm Page 523 Table of contents . and the total safety against creep failure would be JL = 1.step method Comments The agreement between the 1step and 2steps methods is in most cases good.08) and one with a low normal force (nu0 = 0. one could say that the additional “builtin” safety is 0. As shown above.3.118 = 1.32 has been used. and the “safety factor” is JL = 0. A twostep calculation according to the above scheme is done with different values of nL. it is a consequence of the longterm load being close to the instability load for Mef = M. a twostep calculation or a onestep calculation with Mef based on total moments. the result would have been more representative with a somewhat lower longterm load.159.35.4. For the sake of completeness. In this respect. In principle it means that creep deformations will correspond to the stresses in the final stage. until a value is found for which nRd = 0. This is misleading. e02/h = 0.159 / 1. but in these cases the longterm load is close to the limit where instability would occur with Mef = M.36 This is considered to be sufficient. the error will be small.74M The following question now arises: Can load NL together with Mef = M be more severe than ND = 1.g.8.74M = 2. If creep would also be neglected. according to its definition in EN 1990. Thus. Mef = M = 3. the result is in principle already 10% on the unsafe side. The “extra” safety can be estimated by comparison with more accurate calculations.14 1.118 = 1. the results would be another 5 to 30% on the unsafe side. If second order moments are neglected. assuming first order moments proportional to the load as in the above examples. in the extreme case of permanent load only. (The 2steps method in this case seems to give more effect of creep for low than for high normal force.14. 5. There are also a few cases where the opposite is true. The use of an extended stressstrain diagram in the 2steps method can be discussed. Furthermore. although it should be observed that it is not the whole safety factor as the normal material safety factors are already included in the calculated capacities. The conclusion is that creep can not be neglected in the slenderness limit. one case with a high normal force is also included (nu0 = 1.step method nu2 load capacity including the effect of creep according to 2.35 = 0.122. This happens for nL = 0.
This may become the governing factor in cases where the percentage of longterm moment is moderate or high.e.1 General The most accurate of the methods described in 5. \2 = 0. wind. however. including both material and geometric nonlinearity (second order effects). combining analysis and cross section design in one step. If the imperfection is not proportional to the effective length. This is less conservative. together with “stiffness method” for method (a). which is based on the estimation of stiffness.8. since it tells nothing about the method (all methods are based on models). but not in the Eurocodes.8. no other first order effect than the prescribed imperfection. the more variable the load.8. If the imperfection is proportional to the effective length.35 QL and with Mef = M.8.g. However.5 Methods of analysis C5.4 (3) the alternative of using total moments in the definition of effective creep ratio is given. One such method. In 5.8. one diagram or table would only be valid for one length). This is the case in some codes. i. The method rests on a few simple assumptions: • linear strain distribution • equal strains in reinforcement and concrete at the same level • stressstrain relationships for concrete and steel EN 1990 gives values for \2. more precisely when first order moment ratio M0L / M0D > 0. Therefore. this case need not be checked separately. since variable loads are included in QL with \2 Qk. with some simplifications this type of method could be useful also under EN 1992. a separate design for normal force and (magnified) moment must be made.8.8. is proportional to the buckling length of the column.8. Therefore. In the following chapters. the higher the safety against “creep failure” will be. The highest value given is 0. This name will be used here. The limiting factor is the capability of the available computer program. 5. 5. 8 Page 524 Table of contents . It can be used for isolated columns with formally centric load.4 (3) states that a separate check should then be made for 1. 5. A common value is 0.5 is the ”general method”. There are simplified methods other than those mentioned in EC2. whereas in QD they are included with JQ Qk.8. The load bearing capacity is given as NRd = kcfcdAc + ksfydAs (51) where kc and ks are coefficients depending on slenderness ratio. any variation of cross section. any stressstrain relations.6. any boundary conditions. concrete grade. then the absolute value of this length must be added as a separate parameter. calibrated against calculations with the general method.8. since the method is based on the estimation of a curvature.3 (office and residential areas). A special moment magnification factor is included in the method for such cases. e. where JQ > 1.5.can be based on axial loads as well as moments. Therefore. where \2 < 18. axial load and first order moment. and it is not necessary to include any safety factor on M0L in the definition of Mef. which complicates the presentation (for example. In the general case only moments should be used.6. the coefficients can be given in one simple table or diagram with slenderness as the basic parameter. will give sufficient safety against failure under quasipermanent load with full creep. Of the simplified methods. Methods of analysis Three basic methods are described in 5. will be shortly described here as an example (it is currently used in the Swedish code). It is based on nonlinear analysis. For some loads.5. General method 5. effective creep ratio etc. A more suitable name is “curvature method”. but the simplicity is lost and the method no longer has any particular advantages over the “stiffness” or “curvature” methods in EC2. which are common as interior columns in buildings. with some modifications. and most of the extra safety against “creep failure” is then lost. the general method and the simplified methods (a) and (b) are described. The conclusion is that a onestep calculation. imperfection.5. uniaxial or biaxial bending etc. A method of this type works ideally if the imperfection. using an effective creep ratio based on first order moments.6 General method C5. ”General” here refers to the fact that the method can be used for any type of cross section. The old name is not used here. If there are first order moments other than that due to the imperfection. an eccentricity or an initial deflection. particularly for storey high pinended columns. (b) is basically the “model column method” in the ENV.Guide to EC2 Section 5 load then decreases.
in which only one cross section (or certain critical sections) is studied.3. The model in ENV 199211 (Appendix 2) does not comply with this. The safety format defined in 5. The safety format should be compatible with the general design format based on partial safety factors. in cases where failure occurs before reaching the design cross section resistance (stability failure) – unless “critical section” is substituted by some “critical length” (which then remains to be defined.g. in order to avoid discontinuities and computational problems. The reduction of the reinforcement strength is too severe. by using a descending branch of the concrete stressstrain curve in tension.8. A design value of the ultimate load will be obtained as a direct result of the analysis. the simplest way is to multiply all concrete strains by (1+Mef).3. In the calculations presented in this report. but less accuracy. 9 Page 525 Table of contents . e.3). the partial safety factors given for E should be lower than for f: The absolute magnitude can be of importance also in e. 2. as is also the reduction of the material stiffness parameters. where the absolute magnitude of deformations has a direct influence on the ultimate load. It should be possible to use the same set of material parameters in all parts of the member.9 The safety format should satisfy two basic criteria. Creep can be considered in different ways. The safety format is particularly important in second order analysis. since it uses mean values for the analysis and a global safety factor JR = 1.3 and Esm/1. Figure 5. all contributions from concrete in tension have been ignored.4. which primarily deals with high strength concrete according to Swedish rules. however). This may be selfevident. resistance or stiffness. Thus.2 Safety format The safety format in nonlinear analysis has been much debated. See figure 5. it is also convenient for computational reasons.24. but only in the check of rotation capacity. based on using design values in the analysis. Ecm/1. Since the Emoduli vary less than the corresponding strengths. 5. by modifying the stressstrain curve of the reinforcement or by any other suitable model.6.8. This gives the same results as using design values fcm/1. A continuous curve with a descending branch is considered to be the most realistic alternative for the concrete. satisfies both criteria.6. Illustration of accurate (left) and simplified (right) versions of the general method Any stressstrain relations can be used.Guide to EC2 Section 5 Conditions of equilibrium and deformation compatibility are satisfied in a number of cross sections. and the problems associated with the abovementioned safety formats are avoided. since it assumes mean values of material parameters for the calculation of deformations and design values for the check of resistance in critical sections. this is always more or less conservative. having an assumed variation between the selected sections. This gives simpler computer programs and faster calculation. and different models have been proposed.3 to reduce the ultimate load resulting from the analysis.g.24. fyk/1. continuous beams. the contribution from concrete in tension between cracks) can easily be taken into account in the general method. 10 This diagram is taken from [1].6 (6) as a reference for a simplified version. however. Tension stiffening (i. This also means that there will be no “material safety” at all in the calculated resistance. but it is mentioned in 5. and the potential benefits of using a refined method are lost. The model in ENV 19922 (Appendix B) does not comply with this.3. particularly for reinforcement (Esm/1.e. and the deflection is calculated by double integration of the curvature. 1.8. A nonlinear analysis using this safety format will be conservative. see clause 6. and it would normally not have the same direct influence on the ultimate load as in 2nd order analysis. but this makes no difference for what the diagram is intended to show. it makes no difference whether the ultimate load is governed by concrete or steel. and the curvature is preassumed to have a certain variation in other parts of the member.
1 (1. calculated with the general method.15 includes a factor of about 1.25 for one relative eccentricity e0/h = 0.2 and Mef = 0.05 would be logical.e. see clause 4. The most accurate model would be to increase load and time in steps. i. considering the fact that variations in the Emodulus are negligible. Concrete grade is C80. strains would be calculated taking into account their timedependence. For steel. a factor 1.05 for geometrical deviations. Thus. and considering the relationship between strength and Emodulus. see figure 5. Figure 5. n = N / bhfcd.Guide to EC2 Section 5 For concrete. the more the total moment M increases over the first order moment M0. e.1 and different slenderness values O = 35. This nominal second order moment Mu . differences in terms of calculated result are negligible. (Note that the diagram gives axial load and moment in relative terms n and m.8. Assuming a factor 1. m0 = M0/bh2fcd. the analysis can be made either in steps for loads of different duration.8. Js = 1. but also geometrical deviations in the cross section. Interaction curves for columns of different slenderness.) One point on the interaction curve for a given slenderness is obtained by plotting the maximum value of n on the line representing m0 or e0.6. This occurs for O = 105 and 140 in figure 5. figure 5. For each step.26. caused by equal end eccentricities 5.25.3. a reasonable value of the factor for Ec is JcE = 1.2.M0.6. The higher the slenderness.0 has been chosen as a simplification.5 for strength takes into account not only strength variation. a design value Esd=Esm/1. 5. All curves are based on Z = 0. Figure 5. Jc = 1. strains (and corresponding deflections) from the previous step as starting values for the next increment. However. and in order not to deviate from 3.1 for these deviations.26.4 The effect of creep Creep can be taken into account in different ways.M0 is useful as a basis for simplified methods.25. in some cases there is a stability failure before any cross section reaches its ultimate moment.25 show the total moment M as a function of N for a given e0. For creep in slender members in particular. for each step taking the stresses. However.2. Simple way of taking into account creep in general method Page 526 Table of contents . Rectangular cross section. The thin curves in figure 5.M0 between the cross section resistance (curve O = 0) and the first order moment at maximum load represents the second order moment. One such curve shows the maximum first order moment M0 (or eccentricity e0 = M0/N) for a certain axial load N.5 and chapters 7 and 8. This is demonstrated in figure 5. and then the “true” second order moment is less than Mu .g. 105 and 140. A simplified model is to multiply all strain values in the concrete stressstrain function with the factor (1+Mef). see clause 6. where Mef is an effective creep ratio relevant for the load considered.3. 70.5/1. The difference Mu .10 First order moment is constant. see chapter 4. With this model.25. or directly for the design load combination in one step.1)1/3 § 1.3 Interaction diagrams The resistance of slender columns resulting from a general analysis can be shown in a practical form with interaction curves. n and m0 are relative axial force and first order moment respectively.
Other parameters are the same as in figure 5.e. using Mef = 2. this nominal second order moment is sometimes greater than the ”true” second order moment.25.8. there are two principal methods to calculate this nominal second order moment: 1.8. but normally it is reasonable to use one representative value of Mef for a member or even a whole structure. it can give correct end results. For practical design. The relative reduction increases with slenderness. The total moment including second order moment for a simple isolated member is: M M0 M2 M0 N y 1 l2 M0 N r c (61) where (see figure 5.8. i.M0 in figure 5. if given appropriate values. However. Before entering into details of the two methods in chapters 5. as a nominal second order moment. this method is here called curvature method. Interaction curves for Mef = 2. even in cases where the load capacity is governed by a stability failure before reaching the cross section resistance.27.8.25. for Mef = 0.Guide to EC2 Section 5 Figure 5.28) M = total moment M0 = first order moment M2 = second order moment N = axial force y = deflection corresponding to 1/r 1/r = curvature corresponding to y l = length c = factor for curvature distribution Page 527 Table of contents .8.25 are also included (dashed).8.27 is calculated in this way. Curves according to figure 5.8. The difference represents the effect of creep Another question is whether one and the same effective creep ratio should be used along a compression member (or in different parts of a structure).7 2. estimation of the curvature 1/r corresponding to a second order deflection for which the second order moment is calculated. The latter would be the most correct alternative. or if it should vary as the ratio MEqp/MEd may vary. considering geometrical nonlinearity but assuming linear material behaviour).7 and 5. 5. their common basis will be shortly described. Figure 5. Mu . estimation of the flexural stiffness EI to be used in a linear second order analysis (i. see chapter 5.e. this method is here called stiffness method. see chapter 5.5 Simplified methods and their common basis In a simplified calculation method one can use the difference between cross section resistance and first order moment. When this moment is added to the first order moment.25 Dashed curves are the corresponding curves from figure 62. As pointed out above. showing the reduction of the load capacity resulting from creep. a design moment is obtained for which the cross section can be designed with regard to its ultimate resistance.0 and other parameters the same as in figure 5.6.
3 Moment magnification factor C5. The model may also underestimate the curvature in some cases. However. see figure 5. cf. Method based on stiffness 5. Illustration of deformations and moments in a pinended column. (In the figure. First order moment could also be given by eccentricity of the axial load.7.8. taking account of cracking. 5.) The difference between the two methods lies in the formulation of the curvature 1/r.8.g.8. 5. The typical example is where the ultimate load is governed by stability failure.9d (63) This model overestimates the curvature in those cases where yielding is not reached.8.7. In the stiffness method 1/r is expressed in terms of an estimated nominal flexural stiffness EI: 1 M = r EI (62) The stiffness EI should be defined in such a way that ULS cross section design for the total moment M will give an acceptable end result in comparison with the general method. since it does not take into account creep. various corrections can be introduced to improve the result.28. This includes. and M2 can then be written M2 = M0 S2 / c 0 2 2 S EI / l 0 . before reaching the cross section resistance. giving a too conservative end result.1 . the curvature 1/r is estimated directly. The second order moment can be expressed in the following way. first order moment is exemplified as the effect of a transverse load. In the following chapters the two simplified methods will be described and compared to the general method.2 Nominal stiffness 5. creep and nonlinear material properties.7. pinended with a length l = l0.Guide to EC2 Section 5 Figure 5. among other things.8. equation (61) and fig.7 Method based on nominal stiffness 5.7. In the curvature method.1 General 5. This corresponds to c2 = S2.7.8. on the basis of assuming yield strain in tensile and compressive reinforcement: 1 2H yd = r 0.28: 1 l2 M2 = N y = N r c N 2 M l0 § M0 M2 · ¨ ¸ EI c © c0 c2 ¹ (71) With c0 and c2 it is possible to consider different distributions of first and second order moments (primarily the corresponding curvatures).28.1 Basic equations A simple isolated column is considered. e. Solving for M2 gives M2 2 l0 c0EI M0 l2 1 N 0 c 0EI N M0 c2 c0 2 c2EI l 0 N 1 (72) In many cases it is reasonable to assume that the second order moment has a sine shaped distribution.
1 ¹ (74) which corresponds to equation (5.8.7. = S2/c0. parameter taking into account the distribution of first order moment The total moment will be § E · M = M 0 ¨ 1+ ¸ © NB / N . = M0 E NB / N .28) in 5.1 (73) where NB = nominal buckling load (based on nominal stiffness). Page 528 Table of contents .
8.7.1 and NB/N = 2: Comparison between maximum moment according to exact solution and equivalent first order moment (75) with magnification factor (74). this is also consistent with the assumption of a constant equivalent first order moment.4 M01 0.8.3. Equation (74) can then be simplified to M= M0 NB / N .Guide to EC2 Section 5 5.3 (2). which enhances the differences.8.4 M02 This is illustrated in figure 5.2 (2). The example is based on a comparatively high second order effect (N/NB = 0. whereas for c0 = 10 slightly unsafe results may arise.2 Moment distribution In some cases the value of c0 is known. (75) Figure 5. Illustration of equivalent moments in case of differing end moments Equation (74) can be used with the equivalent first order moment according to (75) also. In many cases it is reasonable to assume that first and second order moments have similar distributions. It can be shown that this expression can be used also for structures.30. See 5. in which case § 1. with the wellknown formula for an equivalent constant first order moment: M0e = 0.8. as in the examples mentioned in 5.7.1 (76) This corresponds to equation (5. Therefore c0 = 8 is recommended in 5.7. Slender member with differing end moments according to figure 71 with e02/h = M02/Nh = 0.29.8. Page 529 Table of contents .8.8. Upper thin line = equivalent moment with c0 = 8 Lower thin line = equivalent moment with c0 = 10 Figure 5. An example of the result is shown in figure 72.6 M02 + 0. where two different c0 values were used: 8 and 10 respectively.30) in 5. A reference is made to 5. Thick line = exact solution.3 (2).30 shows good agreement with the exact solution for c0 = 8. The case of differing end moments will be examined more closely.5).29.7. provided a global buckling load can be defined. Figure 5.7.4 for global analysis of structures.
8. for 0. With the “stiffness” method. the maximum first order moment can be expressed on the basis of equation (74). Is = moment of inertia of concrete and steel area The correction factors Kc and Ks can be calibrated using more or less sophisticated models.7.22)) is a more accurate alternative. valid only for reinforcement ratios 0.2 (2) basically two alternative models are given: a) (expr (5.002. assuming that the total moment is equal to the ultimate moment resistance Mu for the normal force N: M0 Mu E 1 NB N 1 Mu E 1 2 S EI l 2N 1 (77) The following is a simple model for the stiffness.8. to give the required agreement between expression (77) and the general method. see chapter 5.26)) is a simplified alternative.01 either method may be used.01 only a) may be used. The values thus obtained can be considered the “correct” ones.6. slenderness ratios. and reinforcement ratios can be determined using the general method. for < 0. on the other hand.3 Estimation of stiffness The maximum first order moment M0 for different axial forces.7. valid for reinforcement ratios down to = 0. expressed as the sum of separate contributions from concrete and reinforcement: EI=Kc Ec Ic + Ks Es Is (78) where Ec. b) (expr (5.002 Ks Kc Ks Kc 1 k1k 2 / 1 Mef . Thus. Es = concrete and steel Emoduli respectively Ic.8. a) if p 0.Guide to EC2 Section 5 5.01. In 5.
01 where 1 0.5Mef .3 / 1 0. (79) b) if p 0.
8. in case the braced units carry a comparatively high vertical load. (710) is the geometrical reinforcement ratio.20 170 where n is the relative axial force. Two different approaches can be distinguished. see chapter 8.8.7.3.4 Linear analysis of structures Clause 5. NEd / (Acfcd) Ac is the area of concrete cross section O is the slenderness ratio. As/Ac Mef is the effective creep ratio.8. even if the geometrical slenderness of individual bracing units is small. (713) The results of calculations with stiffness evaluated according to expressions (79) to (712) are presented in Appendix 2 of this report.7. one based on a magnification factor for bending moments. Without this possibility. It should also be kept in mind that second order effects may be significant in a structure. Page 530 Table of contents . using reduced stiffness(es) taking into account the effect of cracking. the only alternative for second order analysis of structures would be nonlinear analysis.7 opens the possibility of using linear second order analysis for structures.20 More sophisticated models for estimating the stiffness can be found in [2] and [3]. see [1]. the effects of cracking etc. l0/i For cases where O is not defined. Background.3 (criterion for ignoring global second order effects). 5.7.25): k2 = n 0. H.8.30 0. see (712) k1 = fck / 20 (711) (712) O k2 = n d 0. see (711) k2 depends on axial force and slenderness. in the form of comparison with calculations done using the general method. The paragraphs applicable to structures are 5. a simplified alternative to (712) is also given (5.3 (3) and Annex H.2. 5. see chapter 4 k1 depends on concrete strength class. 5. The Appendix also compares the curvature method. When global second order effects are significant. creep and material nonlinearity in a simplified way.3 (3).8. and the other one based on a similar factor for horizontal forces. may be as important as for isolated members.
If properly used.30) and (H. see examples in figure 5. c) H1 is an equivalent horizontal load that would give the same deformation y1.31. The total equivalent horizontal force is H = H0 + H1 + H2 + H3 +… (714) If ki = Hi/Hi1 is < 1.32.. can be used for all kinds of structures. d) Vertical load V and deformation y1 give additional deformation y2.e. a) Horizontal load H0 (without vertical load) gives deformation y0. (No deformation is shown in this case.1. second order effects are instead assumed to be included in the fictitious. see the schematic example in figure 5. 5. Figure 5. on the other hand.8. shear walls with large openings etc. it gives the correct second order effects in structural systems like frames. then the sum H will be finite (i. and it should be used for frames. In other cases second order effects may be calculated stepwise as indicated for a simple frame in figure 5..) Expressions (5. Example of structures where a magnification factor can be applied directly to bending moment(s) in bracing unit(s) The approach based on magnification of horizontal forces.33.g. magnified horizontal force HEd. shear walls with or without openings etc.32.3. the structure is stable).3 and H. Figure 5. With increasing number Page 531 Table of contents . Example of a structure where the magnification factor should be applied to horizontal forces rather than to bending moments. e) H2 is equivalent horizontal load giving the same deformation y2 etc . Figure 5. b) Vertical load V on deformed structure gives additional deformation y1.33. like in certain regular structures. see e. or structures braced by simple cantilever columns.Guide to EC2 Section 5 The two approaches are basically the same.31. Illustration of stepwise calculation of second order effects. but the one based on moments is suitable mainly for structures with bracing units consisting of shear walls without significant global shear deformations..7) are useful if the global buckling load can be defined without difficulty.
like in expression (716).2 are used. the following terms will form a geometric series. The simplest alternative is to assume that all terms.5..34.5. kvalues for early steps would otherwise be too low for later steps. In other words.5 an extended definition of the effective depth d has been introduced. The total equivalent horizontal force is then obtained as H= H0 H0 = 1 .8.n) / (nu . would not be correct in for instance a frame or a shear wall with large openings. including H0.H 1 / H 0 (715) which is equivalent to expression (H. Here correction factors Kr and KM are included: 2H yd 1 1 = Kr KM = Kr KM 0. 1/r is estimated on the basis of reaching yield strain in tensile and compressive reinforcement. Expression (715). However. the relevant second order effects can be obtained everywhere in the structure. the ratio ki will sooner or later become constant. although cracking might occur in later steps. if values for uncracked section are used in early steps. It should be observed that variation in the degree of cracking between first and following steps does not have to be considered. as in expression (76)/(5.8. cf.Effective depth in cross sections with reinforcement distributed in direction of bending In order to reduce the curvature in cases where yielding is not reached in the tensile reinforcement.. the method of stepwise calculation can be seen as just a different formulation or derivation of the method based on a magnification factor.k1 1 .H 2 / H1 1.30).8.8. Thus. To magnify all moments with the same factor. 1 . H = H0 + H1 + . If the distribution of H1 is significantly different from that of H0. Method based on curvature Basic relationships This method is basically the same as the previously socalled “model column” method in the ENV.nbal) 1 Page 532 Table of contents (83) . a factor Kr is introduced (same as K2 in ENV 199211. these values are intended to be valid for the final stage of deformation. etc .Guide to EC2 Section 5 of steps. the accuracy can be improved by including one or more steps: H = H0 + H1 H2 . in order to cover cases where there is no unambiguous definition of d. where V is the total vertical load and VB is the global buckling load. compared to other uncertainties like the effect of stiffness variations within and between members due to cracking etc. can also be expressed in terms of y or a relevant M.6.H 3 / H 2 (716) The simple alternative (715)/(H. When the structure is analysed for the equivalent horizontal force HEd.3): Kr = (nu . It can also be shown that the final value of k is equal to the ratio V/VB. where gradual cracking is taken into account.7. in a linear analysis. including the definition of k.3.34 where is is the radius of gyration of the total reinforcement area Figure 5.9d r r0 (82) In chapter 5.8. Note. see figure 5.8) is sufficiently accurate in most cases. then more steps have to be included in the analysis. equation (61): 1 l2 M2 = N y = N 0 r c (81) As mentioned in 6. 4. if reduced stiffness values according to 5.8). The second order moment is expressed in the following way.8 Method based on nominal curvature C5. This would apply generally when a more refined analysis is used. will form a geometric series. 5.
since the second order effect will make the total eccentricities even “less biaxial”. independent of method.2). This statement is based on [4]. Comparison with general method and stiffness method The result of calculations with curvature according to expressions (81) to (83) is presented in Appendix 2.3.8. see [4].35. given in expression (5. Biaxial bending The general method is suitable for biaxial bending also. 5. 4.8. both fulfilling the criterion for separate checks according to the ENV.9 Biaxial bending C5. which reduces the curvature for values of O between 15 and 35.6.8. If criterion (5.3. “with proper assumptions concerning the distribution of curvature”. however. For details. Thus.8. Simplified methods like the stiffness or curvature method can also be used. whereas in 5. The reason for including the second order effects is illustrated in figure 5.3 (2). the cross section should be designed for biaxial bending.6.4. More sophisticated models for estimating the curvature can be found in [2] and [3].3. A simple Page 533 Table of contents . Example of member with different slenderness in the two directions Assume for example O = 100 in one direction and O = 20 in the other. It has been calibrated against calculations with the general method.5 there is an indication that the curvature method can be used also for structures. They are then used separately for each direction. the factor K1 has not been included in 5. Comparisons with the general method indicate that in certain cases the method can give unsafe results if allowance for creep is not considered.38) is similar to expressions (4.5. since the second order effect will now give total eccentricities outside the “permissible” area. a first order criterion can be misleading and unsafe. The purpose of this factor was presumably to avoid discontinuity in cases where second order effects may be ignored. although the complexity of the problem increases. no further action is necessary. Second order effects will then be significant in one direction but negligible in the other. A and B are two examples of the position of the axial load.75) in the ENV.8. This would be acceptable for case A.9 it concerns total eccentricities including second order effects.9. second order effects will often be ignored for O between 25 and 35 (see 5. and the factor KM has been introduced for this purpose. Furthermore. where a method is given by which the curvature method can be used also for second order analysis and design of unbraced frames. In the same Appendix calculations with the stiffness method (chapter 7) are also presented and compared.35: Figure 5. and if the resulting bending moments fulfil a certain criterion. Their background is presented in [1]. the value 0.8.38) is not fulfilled. The criterion in (5. but there is one important difference: the ENV check concerns only first order eccentricities.74) and (4.4 may be used Z = As fyd / (Ac fcd) As is total area of reinforcement Ac is area of concrete cross section There is another factor K1 in ENV 199211. in comparison with calculations based on the general method. However. It is not acceptable for case B. For these reasons. based on first order eccentricities. there is always a basic discontinuity following from the rule that second order effects may be ignored if they are below a certain limit.38).5. relative normal force ( in the ENV) NEd is design value of normal force nu = 1 + Z nbal is value of n at maximum moment resistance.8. 4. The same principles as in uniaxial bending apply. The ENV gives no indication of how to take into account creep in the “model column” method. so discontinuities will still occur.Guide to EC2 Section 5 where n = NEd / (Ac fcd). Using the curvature method for structures In 5.
3. Verification of new model for slenderness limit Variables covered: Concrete grade: C20. including nominal 2nd order moment MRx/y corresponding moment resistance of cross section a exponent The values of the exponent a are taken from a UK proposal based on [5].11 Analysis for some particular structural members No specific comment on this part. For beams in finished structures.11 Analysis for some particular structural members 5. lateral instability is normally prevented by lateral bracing from adjacent members (e. It is technically equivalent to the corresponding criterion in the new DIN 1045 [8]. the following changes have been made: 1. It is explained and compared to the ENV below. C80 Page 534 Table of contents . 2. Figure 5.40). Lateral instability of slender beams Compared to the ENV. It is quite clear that the ENV criterion does not represent the numerical results very well. floor or roof elements). 5. this is based on national comments. a distinction between persistent and transient design situations has been introduced.36 shows a comparison according to [7] between the numerical results and expression (5. A lateral deflection l/300 has been introduced as an imperfection to be used in calculations concerning lateral instability and balance at supports. It is clearly stated that the check of lateral instability of beams is relevant in situations where lateral bracing is lacking. Expression (5.9 Lateral instability of slender beams C5. “in the absence of an accurate cross section analysis”.40) is based on a numerical study [7]. Criteria for ignoring second order effects in beams according to ENV and EN in comparison with numerical results [7] In the final version. These values can be used in the absence of more accurate values. is given in 6.Guide to EC2 Section 5 model for this.15 5. Figure 5.g.1: § Mx · § My ¨ ¸ ¨ © M Rx ¹ ¨ M Ry © a · ¸ d1 ¸ ¹ a (91) where Mx/y design moment in the respective direction. but it has a different mathematical formulation to show the main parameters l/b and h/b more clearly. The criterion for neglecting second order effects is different. together with an additional criterion for h/b. 5.36. The exponent has been slightly adjusted according to [6]. Appendix 1.9. The DIN model is much better. C40.10 Prestressed members and structures 5. The corresponding criterion according to the ENV is also shown. it is too conservative in many cases and unsafe in other cases.10 Prestressed members and structures See example 6.
3) Effective creep ratio: Mef = 0 and 2 (for C40 and Z = 0.3 also Mef = 1) Explanations to diagrams: Horizontal axis: relative normal force n Vertical axis: slenderness limit lim Full curves: 10%criterion. alternative 1 Dashed curves: 10%criterion.1 and 0. alternative 2 Thick grey curves: new proposal for slenderness limit Concrete C20 Concrete C40 Page 535 Table of contents .Guide to EC2 Section 5 Reinforcement ratio: Z = 0.5 (for C40 also Z = 0.
calculations have been made with the general method according to 5. and it would be difficult to obtain an overall view of the results.7 and 5.Guide to EC2 Section 5 Concrete C80 Appendix 2. The results are summarized in table A21 on the following two pages. In [1] the simplified methods are compared with the general method also for columns with circular cross section and with a different distribution of the first order moment. i. the slenderness. Therefore. However.8. the stiffness and curvature methods according to 5. where s is standard deviation.g. concrete strength fck and effective creep ratio Mef. covering all the abovementioned cases and including the different methods and alternatives. eccentricity e0/h. Calibration of simplified methods A2. the main conclusion is that the same models. Interaction diagrams have also been prepared. for one value of the eccentricity 972/9 = 108 values etc.8. as for rectangular section and constant moment.8 respectively. Thus. to present all these diagrams would require too much space.8.1 Main calculations and results Calculations have been made for isolated columns with the following variables: The total number of individual cases is 4 9 3 3 3 = 972 (not including O = 0 and e0 = 0). can be used also for other cross sections and variations of the first order moment. for one value of e. reinforcement ratio (both Z and ). For each case. The mean value (and standard deviation) for a certain value of the independent variable includes all the values for the other variables. Although [1] deals with more refined versions of the stiffness and curvature methods. The vertical axes in the diagrams represent the ratio Maximum first order moment according to simplified method Maximum first order moment according to general method The moment ratio is given with the mean value m and with m+s and ms respectively. only two such diagrams will be shown to illustrate certain aspects. 12 Page 536 Table of contents .6 and with the simplified methods.e. the mean value and standard deviation of the moment ratio represent 972/4 = 243 individual values. The horizontal axes represent the main variables: slenderness O.
A21.A21. Summary of comparisons between simplified methods and general method Table 5. continued Page 537 Table of contents .Guide to EC2 Section 5 Table 5.
a method giving close agreement with the general method over a wide range of parameter values would no longer be simple.3. This is inevitable. Two sets of curves are given for the curvature method.4.6.24). but becomes extremely conservative for high slenderness ratios. particularly for high slenderness ratios and small eccentricities.37).37). the result is much improved. It is difficult to calibrate a simple method so that it accurately follows the general method.4.2 Discussion All the simplified methods show a rather wide scatter when compared to the general method. Figure A22 shows the corresponding curves for Mef = 2 (a comparatively high value). 4. An illustration is given in figure A21. This is true for all methods. 5.Guide to EC2 Section 5 A2.22) to (5.1. The same is true for the simplest version of the stiffness method (expr. expression (5. The upper curves are based on KM = 1. This is because the factor Kr (K2 in ENV) gives no reduction of the curvature at all when n < 0. see the first and second rows of diagrams in table A21. The lower curves are based on KM according to expression (5. creep is well taken into account. where there is no effect of creep. With KM according to expression (5. Page 538 Table of contents . the curvature method is consistently unsafe for low and moderate slenderness. corresponding to the method in ENV 19921. This can be seen also in the third column of diagrams in table A21.3. With correction of the stiffness for normal force and slenderness. The “curvature method” gives reasonable results for low to moderate slenderness.5. and for high values n is practically always < 0.26). Without effect of creep (= ENV). In figure A21 there is no effect of creep (Mef = 0).
A W and Narayanan. Stockholm. 19990528 Page 539 Table of contents . Handbook published in Sweden. W: Nachweis der Kippstabilität von schlanken Fertigteilträgern aus Stahlbeton und Spannbeton. B: Design Criteria for Reinforced Columns under Axial Load and Biaxial Bending. November 1960. G and Pauli. Mechanics Division. fib (CEBFIP). March 2000 [7] König. September 1999. September 1997. ACI Journal. B: Design Methods for Slender Concrete Columns.und Stahlbetonbau 87 (1992) [8] Hellesland.1. [6] Whittle. [4] Beeby. Beton. Thomas Telford. Tyréns Technical Report 1997:1. 1999. 1995.Guide to EC2 Section 5 References [1] Westerberg. R S: Designers’ Handbook to Eurocode 2. [2] FIP Recommendations. University of Oslo. J: On column slenderness limits. R T and Lawson. An investigation into the use of Bresler coefficients for determining the capacity of reinforced concrete sections under combined axial compression and biaxial bending. Practical Design of Structural Concrete. Part 1. [3] Design Handbook for High Performance Concrete Structures. London. [5] Bresler. R: Biaxial bending with axial compression.
Guide to EC2 Section 5 Page 540 Table of contents .
73625 0.36201 80 0.80952 = 0.39375 60 0.1. Table 6.1.37723 70 0.63719 0.35000 up to 50 0. Values of fck (N/mm2) 1 2 1 and 2 for rectangular diagram 70 0.76781 0.37500 80 0.2.1. respectively t and the formulae b (top and bottom) are given by Ht [ ' Hc2 Hc2 [ ' 1 Hcu 2 § 1· ¨ 1 ¸ Ht [' ¹ © Hb Page 61 Table of contents .35482 90 0.1.36250 90 0. Table 6.³ Vc ( H ) dH The numeric values of 1 and 2 are shown in function of fck in Table 6.41597 55 0.69496 0.56000 0.74194 0.59936 0. the following values result: 1= 2 O· = O /2 the values are shown in Table 6.Guide to EC2 Section 6 SECTION 6 ULTIMATE LIMIT STATES (ULS) 6. In all tables limit the number of decimals to 3 maximum e. are: E1( H cu 2 ) ³ Hcu 2 0 V c dH fcd H cu 2 E2 ( Hcu 2 ) 1 ³ Hcu 2 0 Vc ( H ) H dH Hcu 2 0 Hcu 2 .1 Determining the compression resultant and its position compared to the edge of maximum deformation in case of rectangular section Two cases should be distinguished: e) real neutral axis (x h) f) virtual neutral axis (x > h) Real neutral axis a1) Diagram parabola – exponential – rectangle The resultant C of the block of compressive forces related to a rectangle of width b and depth x is expressed by C= 1 · fcd ·b·x cu2.61625 0. 0. 3.7. where [' = x/h the maximum and minimum strain at the section.1 Bending with or without axial force SECTION 6 ULTIMATE LIMIT STATES (ULS) C6.1 Bending with or without axial force 6. in function of strain c .80952 0.2. is defined by 2 x. 6.58333 0.40000 55 0.39191 60 0.Values of fck (N/mm2) 1 2 1 and 2 up to 50 0.38750 b) Virtual neutral axis b1) Parabola – exponential – rectangle diagram With reference to Fig.35294 a2) Rectangular diagram With the combined effect of the and factors recalled in Chapt.1 .810 etc. and its position. measured starting from the edge where the strain is The formulae of 1 and 2 .67500 0.g.80000 0.
59936 0.75000 0.47385 0.58333 0.(2 .48779 0.80000 0.49705 0.42761 0.48861 0.1.48178 0.49512 0.40186 0.44724 0. k fck (N/mm2) d 50 55 60 70 80 90 0.kh determined by imposing that for x = h is h* = O · h.99060 0. the resultant 3 and its position 4 compared to the most compressed edge and in relation with depth h are given by: E3 [ 'E1t ( [ ' 1) E1b E4 The [ ' 2 E1t E2t ( [ ' 1) E1b (( [ ' 1) E2b 1) E3 3 and 4 values for a number of x/h ratios are given in Table 6.95622 0.85000 0.46046 0.35482 0. It results: 1 k = 2O so that the equivalent depth h* is obtained from the expression: x .73986 0.75000 0.95972 0.91168 0. The resulting values are given in Table 6.90000 0.41022 0.49089 0.87838 0.80952 0.93113 0.80 2.45841 0.96693 0. Page 62 Table of contents .81702 0.50 5.91348 0.88448 0.84129 0.3.69496 0.94460 0.60 1.63719 0.4.96937 0.48347 0.00 1.45389 0.46990 0.36201 0.45954 0.00 x fck = 50 N/mm2 3 4 parabola – rectangle costitutive law fck = 55 N/mm2 fck = 60 N/mm2 fck = 70 N/mm2 3 4 3 4 3 4 fck = 90 N/mm2 3 4 0.37723 0.97481 0.72500 0.93409 0.) h O Values of .96464 0.77500 0.41597 0.5.78714 0. Table 6.49893 0.97500 0.70968 0.47311 0.74194 0.42907 0.40355 0.79679 0.87615 0.80000 k 0.67720 0.91730 0.82826 0.39191 0.45832 0.3.88197 0.4.44335 0.89255 0.95468 0.85234 0.69249 0. k are given in function of fck in Table 6. It is nevertheless possible to write a formula that x .49077 0.46895 0.49057 b2) Rectangular diagram In this case (x h) EC2 does not give instructions.47478 0.45499 0.72968 0. .47702 0.35294 0.47705 0.00 2.73016 0.44461 0.48304 0.82834 0.57143 As an application. Rectangular section with virtual neutral axis Indicating respectively by 1t and 2t the resultant and its position about the whole length x and by 1b and 2b the similar quantities about the xh part.43765 0.46644 0.00000 0.98550 0.84211 0.95000 0.94420 0. Table 6. Values of O. .46140 0.85695 0.78750 0.43492 0.46237 0.49475 0.83288 0.Guide to EC2 Section 6 Figure 6.62069 0. 3 and 4 values fck = 80 N/mm2 3 4 h 1.70000 1.89549 0. so that it results h* = h where the k a factor is x .40 1.42436 0.75381 0.98285 0.44975 0. the values of 3 and of 4 were calculated in analogy to that was developed for case b1).78422 0.20 1.90007 0.78831 0.47480 0.66667 0.99702 0.O h h* = h 1 x .Oh gives the equivalent depth h* in relation with x.49234 0.
74667 0.85909 0.46324 0. On the basis of their level from the stressstrain diagrams of concrete and steel.94341 0.2 Calculation of strength of rectangular section 6.47320 0.88889 0.77677 0.47619 0. 6. 3 and 4 values for rectangular diagram fck = 80 N/mm2 3 4 h 1.80000 0.1.46704 0.45536 0.44767 0.78572 0. In order to determine NRd and MRd two equations of equilibrium (horizontal shift and rotation) are written.98824 0.93554 0.2.67500 0. the design normal force and the design bending moment are determined about the centroidal axis.88269 0.86011 0.89154 0.44444 0.48000 0.00 2.47742 0.42241 0. the reinforcing bars.70000 0.48380 0.81964 0.47225 0.80 2.46154 0.49038 0.94118 0.47727 0.82344 0.40000 0.72800 0.73623 0.46875 0.2) .63636 0.Ns The single terms can be developed as: Nc =b E1 x fcd N’s = V's A’s Ns = Vs As In particular f V's = H's Es if H ' s < yd Es where § d' · H 's H cu ¨ 1 ¸ x¹ © f V's = fyd if H ' s t yd Es likewise f Vs = Hs Es if H s < yd Es where §d · Hs H cu ¨ 1 ¸ ©x ¹ f Vs = fyd if H s t yd Es The design bending moment about the centroidal axis is expressed as Page 63 Table of contents (6.75938 0.45720 0.73625 0.47004 0. Equilibrium to shift: if Nc is the resultant of compressive stresses applied to concrete.91073 0.40997 0.79773 0.92274 0.69695 0.45577 0.48571 0.83382 0. the materials and the line that defines the deformed configuration at ultimate limit states.38750 0.44674 0.35000 0.00 x fck = 50 N/mm2 3 4 Rectangular costitutive law fck = 55 N/mm2 fck = 60 N/mm2 fck = 70 N/mm2 3 4 3 4 3 4 fck = 90 N/mm2 3 4 0.89308 0.48548 6.40 1.96000 0.85601 0.47469 0.77482 0.1.42188 0.71628 0.49329 0.93097 0.75946 0.46332 0.1 Determination of NRd and MRd Given a transverse rectangular section with symmetrical geometry and reinfocement.5.56000 0.46219 0.46667 0.Guide to EC2 Section 6 Table 6.20 1.48809 0. the corresponding stresses are calculated.49239 0.96191 0. Rectangular section at ultimate limit state On the hypothesis that straight sections remain straight.43339 0.43308 0.88030 0.95238 0.36250 0.00 1.82975 0. deformation are as in fig.80282 0.84375 0.91534 0. Figure 6.39773 0.60 1.90191 0.2.39375 0.67586 0.45500 0.2.44318 0.97143 0.48176 0.50 5.45269 0.43898 0.76781 0. N’s the resultant of stresses applied to the compressed reinforcing bars A’s and Ns the resultant of traction in the reinforcing bars As .61625 0. NRd = Nc + N’s .47059 0.43750 0.1) (6.37500 0.92308 0.49412 0.
Note The procedure exposed in the general guidelines follows from the principle of committing. The cu2 (for parabola and exponential – rectangle diagram) and cu3 (for bilateral and uniform) values are identical [Table 3.Guide to EC2 Section 6 h h h Nc ( E2 x) A 's Vs' ( d ') A s Vs ( d ') . The yd value depends on the steel design stress fyd = fykJs .Mlim is to be absorbed by two sets of reinforcement. Page 64 Table of contents . More severe limitations of x than those above. which both work at the design limit of elasticity.3) and load effects in the plane of symmetry.d' ) The area of reinforcement steel. 2 2 2 In case both the reinforcing bars are yielded (Vs = V's = fyd ). one compressed and one tensioned. If MEsd is greater than the moment Mlim. of area MEsd . In such cases. Design of reinforcing bars in case of bending without axial force and in case of bending with great eccentricity axial force Let's take a transversal section with axis of symmetry y (Fig. other limitations apply (see note). This limitations is valid for isostatic members.: M Rd NRd = b x fcd . where redistribution of moments may take place. the strain s of the tensioned steel must be greater that the strain corresponding to the limit of elasticity. and only in case the tensioned steel is not sufficient.3. the bending moment about the tension reinforcement is calculated: MEsd = MEd . as far as possible. compression to concrete and traction to steel.2. A s' = fyd (d .are required in order to meet ductility requirements in the case of indeterminate structures in bending. A’s . In order to ensure that the structure has a ductile behaviour.A's fyd + As fyd h M Rd E1xbfcd ( E2 x) (A s A 's )f yd (d d ') 2 1 and 2 are factors given in Tables 6. that corresponds to xlim in presence of tensioned reinforcement only. a certain amount of compressed steel has to be put in place. for other cases. The difference MEsd = MEsd . 6.NEd · ys and the reinforcement area As required by MEsd is calculated like in the case of simple bending moment.3. The value of cu depends exclusively on the concrete class.1 or 6. that is s yd = fyd/Es . Given the design load effects at ultimate limit state MEd and NEd. has to be added to the section As corresponding to the limit bending moment.1EC2]. the applied axial force must be taken into account with its sign: if NEd is a compression force. 1 (6. it will reduce the area of tensioned steel.3) 6. This implies that the neutral axis does not exceed the depth H cu d x lim H cu H yd where cu is the strain at the compressed end. Figure 6. The axial force NEd is taken into account subsequently. by correcting of an equivalent quantity the resistance of the tensioned reinforcement steel. some compressed steel (A's ) is added.1. Simple bending moment and composite bending moment with great eccentricity A general rule is adopted: only the tensioned steel (As ) is provided.
1. the bending moment (Fig. developed. both in statically indeterminate and determinate cases.MRd. Given the section dimensions (b. h. In order to determine it the value of x corresponding to MEsd must be defined. the resultant of traction of the reinforcement steel As .3) about the tensioned reinforcement elements is calculated: MEsd = MEd – NEd .1 Use of parabolarectangle and exponentialrectangle stressstrain relations Design is simple if the 1 and 2 values (respectively the resultant and its distance from the edge. In the design process point [9. the verification of serviceability tensional stresses [7. 2 Remembering that MEsd = As fyd z. with z = (d finally As M Esd f yd z As Fc must be equal and contrary to Ft . it's necessary that the resisting bending moment of the reinforced section is greater than the moment that causes cracking.lim . d. As alone is needed.lim = Fc·zlim where Fc = arm. 6. In other words.1. effective depth defined as distance of the tensioned reinforcement steel centroid from the compressed edge). MEsd = MEsd .lim with tensioned reinforcement only is calculated MRd. for an element of unitary width and depth).2.1. from which: 'M Esd A 's f (d d ') yd The tensioned reinforcement is: As 1 f yd §M · ¨ Rd . viz.lim .2(2)EC2] implicitly requires that the depth of neutral axis at ultimate limit state is limited.4 Rectangular section 6.2.lim. it can also be determined by: As = 1·b·x ·fcd/fyd If MEsd is greater than MRd.ys . To calculate it. It results: 1·b·x ·fcd· (d 2 ·x) = MEsd which. are used. materials and action effects.Guide to EC2 Section 6 Moreover. Page 65 Table of contents . the following procedure is followed: xlim is determined the limit bending moment MRd. some reinforcement steel A's in compression is needed. In order to determine if As is sufficient. 6.lim N Ed ¸ A 's z lim © ¹ In such cases MEsd must be sensibly smaller than MRd. 1·b·xlim·fcd is the resultant of compression stresses and zlim = (d 2 ·xlim) is the inner lever a) If MEsd is smaller than Mrd. or if also A's is necessary. it must not distort the problem.1(1EC2] must also be taken into account requiring a minimal quantity of tensioned reinforcement steel to avoid that fragility situations arise side steel.lim.4. given at point 7. becomes: x2 x d M Esd E2 E1 E2 b fcd 0 Solving: x § d · d M Esd ¨ ¸ 2E2 2 E2 ¹ E1 E2 b fcd © 2 ·x).
3.y/2) Figure 6.4. a lower bulb for the placement of tensioned reinforcement steel could be needed. about the same layer. M* is the given bending moment about the As reinforcement centroid. the lower layer is also surely yielded. It results: Fc= b·y· ·fcd z = (d – y/2) MRd = Fc ·z = b·y· ·fcd ·(d . Values of and are given in Table 2. the considerations and developments of the previous point apply. the equilibrium to rotation between external and internal moment is written with reference to the layer that corresponds to half way between the reinforcement layers As (steel reinforcement centroid). each one of area As. The upper layer will have distance d from the compressed edge so that. as this reinforcement layer will have at least strain yd .4.4. the tensioned reinforcement steel should be laid on two layers. the reinforcement elements are determined by equilibrium to shifting.3. The bending moment resistance is expressed as the sum of the moments of blocks of compression. Values of the O and K factors are given in Table 6. that is used to determine As. 6.7(3)EC2] is adopted. especially if medium/high strength concrete is used. also contain NEd.Guide to EC2 Section 6 6. An approximate method is also presented.2 Rectangular diagram of concrete stresses With reference to Fig. The rectangular stress diagram is adopted for calculation.1.1. The block of compressive stresses is defined by a uniform value fcd and by the extension y = Ox where O and are two factors lower than 1 and function of fck according to the formulae [3. 6. The equilibrium gives: Page 66 Table of contents .22EC2].5 y is assumed as the basic parameter. In case of bending with axial force the reference layer of bending moments M* (sum of the given moment and of the one deriving from the shifting of the force NEd) will still be the aboveindicated one. 6. The position of the neutral axis is determined through this equation. 6. In case of simple bending moment.d 2 ( Kfcd ) © © 2 0 with the solution y d 1 1 2 M Esd b d 2 ( K fcd ) Also in this case.4. Rectangular section with rectangular concrete stress diagram Rearranging: 2M Esd §y· §y· ¨ ¸ 2¨ ¸ d¹ d ¹ b.1 General method Introduction As very high values of resisting moments can be reached with Tsections. Procedure With reference to Fig.4.1. In this way the contributions of the two tensioned reinforcement layers do not appear in the equation. the neutral axis crosses the web: its determination and the subsequent developments are simple if the rectangular stress diagram provided for at paragraph [3.4.1. In some cases. Then.3 Tsections Two situations can arise in Tsections: the neutral axis is in the flange: no difference as if the section was rectangular. y = O·x is the depth of the compressed zone and ·fcd is the design tensile stress. In this case the equilibrium to shifting. s is the distance of this centroid from the concrete compressed edge.19 to 3.
b w .Guide to EC2 Section 6 ª § y· § c ·º fcd «b w y ¨ s .¸ + b .
2sy + 2M * bw § c· b . c ¨ s . it results: y 2 .bw .¸ » = M * © 2¹ © 2 ¹¼ ¬ By developing.
¸ = 0 fcd . c ¨ s .
Tsection treated with the rectangular stress diagram Taking: k= § c· b . © 2¹ Figure 6.bw .5.
¸ bw © 2¹ 2 with k function of the geometrical data only. y is given by: 2M * y = s .s2 +k b w fcd . c ¨ s .
It's assumed that tensile stresses are uniformly distributed. The tensioned reinforcement area is given by: 2As = (Fc +NEd )/fyd and the resultant of compression Fc is: Fc = K fcd·[bw·y + (b – bw)·c] If it is not necessary to put the tensioned reinforcement bars on two layers.2 Approximate design method The method applies to those Tbeams where the flange is able to withstand all compressive forces deriving from bending moment and axial force.4. the total bending moment M* about the centroid of tensioned reinforcement bars is calculated: M* = MEd . in the aboveshown formulae is is sufficient to identify s as d.6. 6. no compressed reinforcement bars are needed. With reference to Fig.3. Tsection.6. If M* is lower than the limit moment. without addition of compressed reinforcement bars.1.¸ © 2¹ Page 67 Table of contents . Fig.6.13) gives the necessary area of reinforcement. while the first member of (7. 6. Approximate design method The reinforcement is given by the formula ª º » 1 « M* 2A s = « + NEd » fyd « § c · » s«¨ 2 ¸ » ¹ ¬© ¼ and it must be verified that the average compressive stress in the flange is not greater than fcd: M* = d fcd c § c· bc¨s .y*· NEd (NEd positive if traction).
7. The analysis is carried out through the determination of the moment resistance MRd about the centroidal axis for given values of NEd and geometric and mechanical properties of the section.71 3. Four deformed configurations (1234) are taken into account. the straight line assumes configuration 1 characterized. calculated by (6. for x x1 it works at stress fyd. are characterized by particular depths of the neutral axis x. is vertical in the representation and passes by point C. Strain of upper reinforcement A's (see point 6. Table 6.61 0. 6.64 0. is function of fck ).27 2. The lower reinforcement layer. by strain Hs Hsyd J s Es with tyk = 450 N/mm2 and Es = 200 kN/mm2). Fig.0035 0.57 0. works at stress fyd . The depth of the neutral axis. that configures uniform strain c2.6 values of k1 and k2 fck (N/mm2) d 50 55 60 70 80 90 cu2 k1 2.5 Bending with axial force in rectangular section with symmetric reinforcement bars The ultimate limit state behaviour of a rectangular section with reinforcement bars As = A's placed symmetrically at the section edges and subjected to simple bending moment with an eccentric axial force NEd is studied. Possible strain distributions at the ultimate limit state Relations that are recalled in the treatment:.7. obtained by rotation around point B of the straight line that represents the plane section.06 k2 0.0027 0.2.2. line 4. are given in Table 6. The first three.0026 Line 2: it is defined by the strain s = syd of the lower layer of reinforcement.0029 0.0031 0. for x x1 .6 For x < x1 the upper reinforcement layer is in elastic field. In this case the depth of the neutral axis. deduced by (6. The study starts from Fig.Guide to EC2 Section 6 6. 00196 (design limit of elasticity for B450 steel reinforcement layer.6.1): Hs Definition of line 1 Passing by B.08 3.1.1.65 4.57 0.1.1): Hs ' § d' · Hcu 2 ¨ 1 ¸ x¹ © §d · H cu 2 ¨ 1 ¸ ©x ¹ Strain of lower reinforcement As (see point 6.58 0.0026 0.59 0. 6.2) results: Page 68 Table of contents .1) results: x1 § Hcu 2 ¨ ¨ H cu 2 Hsyd © · ¸ d ' k 1d ' ¸ ¹ cu2 The k1 values for different concrete classes (as ( cu2 in absolute value). at the level of the upper f yk ' 0.06 4.1 which is reproduced as far as it's needed in Fig.
it results: § N + A s fyd . The following values of the axial force resistance NRd correspond to configurations 1. with strain s syd. Therefore in the 34 range stress is s = fyd for the upper reinforcement and s = s·Es for the lower. In configuration 4: s = s' = c2 (in absolute value). * * . the lower reinforcement at stress increasing from 0 to fyd. that is x < x1 The position x of the neutral axis must be preliminary determined by the equation of equilibrium to shifting. Configuration 3 is characterized by strain value s = 0 (and therefore s = 0) for the lower reinforcement (x = d). written in synthesis x2 . the equilibrium is written as: . NRd3 = 1bdfcd + A'sfyd NRd4 = bhfcd +2Asfyd.3.4: NRd1 = 1bx1fcd NRd2 = 1bx2fcd In these two cases there are no contributions from the reinforcement bars because these are subjected to ±fyd. 1 values are given in Table 6. In the transition from 3 to 4 the upper reinforcement is compressed at stress fyd. For positions of the neutral axis x1 x x2. Keeping in mind that the upper reinforcement A's is compressed in elastic field and that the lower reinforcement As is tensioned at stress fyd. It results then s = s' = fyd .2 for the rectangle model.2.¨ Ed ¸x¨ 1 b fcd © ¹ © The equation.'s A s . both reinforcement layers are subjected to a stress s = fyd (compression for the upper one.cu2Es A s · § x 2 . Line 3' is defined by c= 0 at the lower edge of section.1 from the parabolaexponential rectangle model and in Table 6. and generate two equal and opposite forces in equilibrium. Calculation of the moment resistance in the four abovedefined sectors.6.1x b fcd + A s fyd = . traction for the lower one).1).NEd Taken 's =E 's where 's is given by (6. a) NEd < NRd1 .Guide to EC2 Section 6 x2 § Hcu 2 ¨ ¨ Hcu 2 Hsyd © · ¸d ¸ ¹ k2 d k2 values are shown in Table 6. which is always greater than syd.
x . * * * .
NEd is exclusively supported by concrete.d') + A s s '( . = 0 has the solution 1 1 2 x = + (**) + (**) + (* * *) 2 4 cu2 s E A s d' · ¸=0 1 b fcd ¹ § d' · ¨ 1.2 x) 2 2 2 The 's stress is now known and adds up to ' s = Es cu2 b) NRd1 NEd NRd2 As both reinforcements are yielded. The equilibrium equation is: NEd = NRd = 1b x fcd . so that the depth of the neutral axis is: x= NEd 1 b fcd Page 69 Table of contents .d') + 1x b fcd ( .¸ © x¹ and the moment resistance about the centroidal level is: h h h MRd = A s fyd ( .
Guide to EC2 Section 6 With the abovedetermined x value. the moment resistance is: MRd = A s fyd (d .2 x .d') + 1x b fcd 0.5h .
such as those of columns. 1) given in Table 6. which shows the final end of the interaction diagram MN.05000 (bottom reinf.0035 at upper end c= .5/(3.6 Interaction diagram MRdNRd In the case of sections subjected to bending with axial force with small eccentricity.) c= 0.2) Replacing and developing it results: § NEd . with 7 configurations characterized by strain values of plane section (reference to Fig. in the four conditions: no reinforcement symmetric reinforcement (fyk = 450 N/mm2 ) at the edges in percentage 0.0020 everywhere s= s= + 0.6 mm Page 610 Table of contents .02500 (bottom reinf. s = EsHs e Hs is given by (6. the most logical solution is the one with double symmetric reinforcement. distanced at d' =50 mm.0. Fs = +939 kN Neutral axis: x/d = 3. by associating the parabolarectangle diagrams for concrete and the bilinear diagram or steel.6).) s= 0. F's = 939 kN Lower reinforcement: s = + 0.) s= + 0.0035 at upper end c= . Such sections can also withstand simple bending and.0035 at upper end c= .00000 (bottom reinf.1x b fcd + A s s (6.0.9) with dimension h = 600 mm.010 and therefore s = + 391 N/mm2 . Table 6.5 on each edge.0020 everywhere Development of the calculation in relation with the third strain condition for the reinforced section with 1% bilateral reinforcement (As = A's = 2400 mm2 ) Upper reinforcement: 's = .00000 at lower end c = . for each of the four reinforcement conditions. 6. Terminal end of the interaction diagram MN The moment resistance reaches a maximum for x = x2 where the analytic function that expresses it has an edge point due to the discontinuity between (6. b = 400 mm of fck 30 concrete. The calculation of the resistance of this section at the ultimate limit state is developed. x = 142. Steel B450 (fyk = 450 N/mm2).A ' s fyd .01000 (bottom reinf.cu2Es A s · § cu2Es A s d · x2 . composed bending in relation if the dimensions and placement of reinforcement.0023 and therefore 's = 391 N/mm2.0.0.6) d) In the fourth field (NRd3 d NEd d NRd4 ) the moment resistance can be determined. the moment resistance results: h h h M Rd A 's f yd ( d ') A s Vs ( d ') E1xbfcd ( E2 x) ( 2 2 2 6.5 – 1.5+10) . The derivative for x = x2 is positive if (6.0035 at upper end c= .7.5) As the reinforcement As is in elastic field.0.¨ ¸x¨ ¸=0 ¨ ¸ 1 b fcd © ¹ © 1 b fcd ¹ (6.4) is used and is negative with (6.0.) + 0.00196 (bottom reinf.4) = NEd (6. and as d = 550 mm. if it's the case.8. Figure 6.4) and (6. let's consider the load capacity of a rectangular section (Fig.7.8. by the relation of proportionality indicated in fig.0.1. 6.0035 at upper end c= . 6.0. Deformed section configurations c= . In order to have an overview on the problem. with a good approximation.6).5bis) Once determined x e s.) s= + 0.0035 at upper end c= .0 – 1.0. c) NRd2 d NEd d NRd3 The equation of equilibrium to shifting is: fyd A ' s .
where the bending components lay on the two centroid axis of inertia.1(4)EC2] a minimal eccentricity must always be taken into account. 6. and calculation developed by hand. For each value of the percentage of reinforcement the points are joined by a straight line. and to two bending components MEyd e MEzd expressed by two vectors orientated along a couple of orthogonal axis y e z with origin in the centre of gravity. according to [6. 6. adding up to the bigger value between h/30 and 20 mm. 5. as it's possible to verify by tracing a line parallel to the M axis. For pure biaxial bending. Given a section subjected to an axial force NEd applied on the centre of gravity. the points on the polygon strictly verify the ultimate limit state.7 Biaxial bending and bending with axial force Biaxial bending may be separated into separate uniaxial bending components under circumstances laid down in Eurocode 2 5.9 . the existence of a stress distribution that gives place to resistance greater or equal to the action effects must be demonstrated. This means that with a given reinforcement.5 = 658. The polygon related to the nonreinforced section denotes the possibility to withstand bending moments only if they come with an adequate axial force (provided by self weight or by prestressing). If a design software is not used. 7. The straight line of inclination M/N = h/30 = 20 mm defines the field of use of the polygons: in fact. the problem solution has computational difficulties.6 kNm The eccentricity of NRd is MRd / NRd = 839 mm. The following observations arise from the observation of Fig.9: 3. The limit case of a single N value happens when the straight line passes by the highest point of the polygon. The result is convex polygons. NEd placed inside the polygon are in a safe zone. it should be processed by iterations. Fig. 6. Such polygons as those traced below are called “MN interaction diagrams”.617.8095400142. Interaction diagram for rectangular section 6.0 = 785 kN Distance of Fc from the compressed edge: 2x = 59 mm NRd = Fc + Fs + F's = 785 – 939 + 939 = 785 kN Moment about the concrete section centroid MRd = Fc (d/2. In general.9 graphically shows the 29 pairs of results obtained.1. in such case it's convenient to adopt the rectangular stress diagram for concrete. external point do not meet resistance conditions at ultimate limit state 4. The branches on the left of the M axis denote resistance to bending with positive axial force 8. a straight line parallel to the N axis intersects the polygon in two points. This implies that the eccentric axial force must also lay on that line. a given bending moment can be withstood by two different values of the axial force.Guide to EC2 Section 6 Fc = 1bxfcd = 0. the polygons include the domains of resistance: the points of coordinates MEd. Page 611 Table of contents .1 + 469. as well as the fact that the straight line that connects the centre of gravity of the compressed zone with the centre of gravity of the tensioned reinforcement is perpendicular to the resultant bending vector MEd.9.2x ) + Fs (d – d’)= 189.8. Figure 6. Only one value of M can be associated to a given value of N.
2 Shear capacity of members without shear reinforcement 6. 6. in the area where the member is not cracked in flexure. k factor allowing for the size effect.7) Where WRd basic shear strength.2. It is necessary to make a distinction between shear flexure and shear tension. Such members are for instance prestressed hollow core slabs. longitudinal reinforcement ratio and crosssectional height was 1/3 (6. which follows from WRd = 0. 1987.02 l design axial stress (if any) = NSd/Ac cp bw minimum web section The are two shortcomings with regard to the use of this equation.2. In this chapter only shear flexure is regarded. which was believed to take appropriate account of the most important influencing factors like concrete strength.2.05/Jc. As/bwd < 0.0.2 Members not requiring design shear reinforcement C6.25fctk.1. The second problem is that the equation has principally been derived for beams. which can be considered as the general case. The basic formula is given as Eq. The recommendations for the determination of the shear flexure capacity of members not reinforced in shear are given in chapter 6. as was demonstrated in [Walraven. but if the strength increases the deviations soon reach an unacceptable level.Guide to EC2 Section 6 6. pp.2.15 cp]bwd (6. 68 .7 would give unnecessary conservative results. failing in shear flexure and is not valid for members which typically fail in shear tension. The basic equation adopted.6 – d (m) > 1 flexural tensile reinforcement ratio.2.2 + 40 l) + 0.] For lower strength concrete classes the deviations were not yet very large. At first the role of the concrete strength is not correct.a in this document. which nearly always fail in shear tension. Applied to such members Eq. In ENV 199211 the equation for the shear capacity of members without shear reinforcement was VRd1 = [WRd k (1.8) Vu = C k 100Ul fc .2 Shear 6.71. 6. This equation has been derived in the following way.2 of prEN 199211:2001. Shear flexure capacity Most shear failures occur in the region of the member cracked in flexure. equal to k = 1.2.
b w d where k = size factor = 1 + (200/d)1/2 = longitudinal reinforcement ratio l fc = concrete cylinder strength (N/mm2) C = coefficient to be determined A selection was made of a representative number of shear tests, considering a parameter variation as wide as possible and as well as possible distributed within practical limits. This was already done by König and Fischer (1995). An overview of the test parameters is given in Fig. 6.10. Then for every test result the optimum value C was determined. If the distribution is normal, Fig. 6.11 a lower bound value for C was determined according to the level 2 method described in [Taerwe, 1993 ] with the equation: Clower bound = Cmean (1D E Q) (6.9)
where D sensitivity factor, equal to 0.8 for the case of one dominating variable (concrete strength) reliability index, taken equal to 3.8 according to [Eurocode, Basis of Structural design, Draft version 2001 ] Q standard deviation If the distribution turns out to be lognormal, Fig. 6.11, the equation is Clower bound = Cmean exp (DEQ Q ) In these equations a reliability index = 3.8 means a probability of occurrence of 0.0072%. König and Fisher (1995) carried out this procedure for 176 shear tests. (6.10)
Page 612
Table of contents
Guide to EC2
Section 6
Figure 6.10. Relative frequency of parameters in test data bank used by König and Fischer (1995) in order to find a reliable lower bound equation for the shear capacity of members without shear reinforcement
Figure 6.11. Normal and lognormal distribution
As a result of their analysis they found that a coefficient C = 0.12 would be a good lower bound. In Fig. 6.12 it is shown that the prediction accuracy of this equation is substantially better than that of the old EC199211 formula.
Figure 6.12.
a. Shear capacity according to Eq. 6.8 (MC 90): relative frequency for NSC and HSC (König, Fischer, 1993) b. Shear capacity according to Eq. 6.7 (ENV 199211): relative frequency for NSC and HSC, according to König, Fischer (1993)
As an addition Fig. 6.13 shows an evaluation carried out by Regan [Regan, 1993], which confirms the findings by König/Fischer.
Page 613 Table of contents
Guide to EC2
Section 6
Figure 6.13. Shear strength of nonprestressed members without shear reinforcement, comparison of test results with Eq. 6.8 [Regan, 1999.]
It was however argued, that the equation VRd,c = 0,12 k (100 Ul fck)1/3 bw d (6.11)
has two disadvantages the first is that it does not distinguish between persistent & transient loading combinations and accidental loading combinations, for which different safety levels apply (prEN 199211:2001 chapter 2.4.1.4 gives Jc = 1,5 for persistent and transient and Jc = 1,2 for accidental situations). Therefore the equation was modified by introducing the concrete safety factor explicitly. VRd,c = (0,18/Jc) k (100 Ul fck)1/3 bw d (6.12)
The second is that the shear capacity goes to 0 when l = 0. Furthermore it was wished to have a simple conservative value for VRd,c for a first check of the bearing capacity. In many countries simple formulations have been used on the basis of VRd,c = C fctd bw d (6.13)
where fctd is the design tensile strength of the concrete and C is a coefficient. Practice in the various countries however is quite different because C varies in the range from 0,3 to 0,75. Considering the value of C it should be noted that this equation is a simplification of the rigorous one. To have general validity, even for rare but still possible cases, C should be based on the most unfavourable combination of parameters. That means that the governing case is a slab with a large crosssectional depth d and a low longitudinal reinforcement ratio. In his paper “Basic facts concerning shear failure”, Kani (1966) showed that shear failures are unlikely to occur for longitudinal reinforcement ratio’s smaller than 0,6%. However, his “shear valley” was based on beams with a crosssectional effective depth of only d = 270 mm. For larger depths the critical value of 0 decreases. Therefore a number of shear failures reported in literature have been selected with large d and small 0 values, see Table 6.8.
Page 614
Table of contents
this method works well for the evaluation of laboratory tests but is less suitable for real Page 615 Table of contents . Some questions may be raised with regard to the definition of bw being “the smallest width of the crosssection in the tensile area”.14. left (tests by Leung. The equations in prEN 199111:2001 contain as well a term 0. A possible compromise could be to define bw as the average width of the part of the crosssection in tension. 6.14.35 would be appropriate for the simplified design equation. with d = 500 and 750 mm and 0 = 0. So. Calculation of contribution Vp from prestressing to the shear resistance according to Hedman and Losberg (1978) However. Fig. with a high crosssection. Fig. 1976) In a more recent publication (Regan. 1976). a critically low reinforcement ratio and a line load just at the most critical position from the support would then have a slightly lower safety. a prestressed beam can be regarded as a reinforced beam after the decompression moment has been reached. subjected to uniform loading. An argument might be that the utmost part of the practical cases consists of slabs with smaller depths. which would be an argument in favour of the use of 0.11 and 16.46 and 0.15. Figure 6.40 is used. Determination of C on the basis of selected tests The most unfavourable values for C are 0. On the basis of this argument the shear resistance was formulated as VRd. It was argued that. with regard to the behaviour in shear. Chew and Regan. Chew and Regan. Shear resistance of beams with tapered crosssection (Leung.15. and the reinforcement ratio’s are small enough to ensure failure by bending. 6.8. In prEN 199211:2001 a value 0.Guide to EC2 Section 6 Table 6. Basically the influence of prestressing can be taken into account as proposed by Hedman & Losberg (1978). 6. with a maximum of 1.15 cp regarding the influence of an axial force on the shear capacity. which can be formulated as Vp = M0/a. found for Aster&Koch’s tests Nr. shows that a definition of bw as the average width of the beam would be appropriate for this case. as shown in Fig. Figure 6. Tests on tapered crosssections showed that there is certainly an influence of the definition of the web width.35. for instance by prestressing.25 of the minimum width. with some rounding off a value C = 0. where the maximum shear force does not coincide with the maximum moment.14. right. 2000) the author opts for a definition of bw = 2/3 bmin + 1/3 bmax. The seldom case of a slab spanning in one direction.42% respectively. but admits at the same time that the available evidence is rather scarce.34.c = Vc + Vp where Vc is the shear resistance of a similar nonprestressed beam and Vp is the contribution of the prestressing force to the shear capacity. where M0 is the decompression moment and a is the distance from the load to the support. On the other hand formula’s should always be safe enough to take account of any possible (not likely) case.
Beams with a rectangular crosssection were provided with nibs. The force could be applied in two ways: before subjecting the member to transverse loading. where Mx and Vx are the bending moment and the shear force in the section considered. or in proportion to the transverse loading.14) In most tests on shear critical beams the ratio ep/h is about 0.0. The effect of longitudinal compression should.16.85h this would result in. For axial tension in prEN 199211 the same formula is used. 1999] For similar arguments. With a/d varying between 2. 6. Nielsen (1990) compared the shear equation in ENV 199211 which gives about the same results as Eq. with a different sign for 0. Vp = 1. It can simply be derived that for a rectangular crosssection with a width b.18 Fp (1/6 + ep/h) (a/d) (6. This was for instance shown by Regan [Regan. This effect. It should be noted that in continuous beams there is tension in both top and bottom and excessive curtailment at sections of contra flexure may lead to diagonal cracking and shear failure in such a region.15 cp. 6.25 cpb d. This was the main cause of failure in an actual structure [Hognestadt and Elstner. not be mixed up with the effect of the cable curvature.16. of course. and failing in shear [Regan. 1971 and 1999] who carried out a systematic investigation into the effect of an axial tensile force on the shear capacity of both members unreinforced and reinforced for shear.15 cpb d and 0. Another disadvantage is that Vp would go to infinity in a moment inflexion point. this would complicate the shear design because then Vp would be different in any crosssection. although the member sometimes showed wide open cracks across the total cross section in the moment inflexion region.35. a height h and an eccentricity of the prestressing force ep. which exerts a favourable transverse load on the member. 1957]. where Mx = 0. this would mean that Vp would vary between 0.2a in prEN 199211:2001 for moderate concrete strengths. enabling the transmission of an axial tensile force in the middle part. bending and shear. with 287 test results and found that it was at the safe side.Guide to EC2 Section 6 members mostly subjected to uniformly distributed loading. is introduced as a load (load balancing principle). The axial tensile force varied between 0 and 130 kN. so that an axial tensile force gives rise to a slight reduction of the shear capacity. Tests have been carried out according to the principle shown in Fig. Evaluating test results it is therefore not amazing that the coefficient 0. In both cases the shear capacity was hardly influenced. Results of tests on beams subjected to axial tension.15 turns out to be a safe lower bound in shear critical regions.5 and 4. the contribution Vp to the shear resistance is Vp = Fp (1/6 + ep/h) (a/h) Assuming d = 0. However. A solution is to replace M0/a by M0/(Mx/Vx). Figure 6.15) (6. If a structural member is well designed for axial tension the shear capacity of the members is hardly reduced. like in most shear tests. known as the load balancing effect. reference is made to [Bhide and Collins. 1989] Page 616 Table of contents .
2. such as in prestressed hollow core slabs. is equal to 1 § 2 1 2· (6. Fig.16) I =N+ N¸ ¨W + 2 4 ¹ © substituting W = VRd.17. shear tension failures can occur.18.18. like for instance when pretensioned strands are used in members with reduced web widths.2.ct = w fctd . The principal tensile strength in the web calculated using Mohr’s circle. Eq.17. 6. 6. Shear tension failure Figure 6. 6. Figure 6.ct S/bwI and N = DI cp the code’s expression EC2. Calculation of shear tension capacity with Mohr’s circle In this case failure occurs due to the fact that the principal tensile stress in the web reaches the tensile strength of the concrete in the region uncracked in flexure.2 Shear tension capacity In special cases.Guide to EC2 Section 6 6.3 Ib 2 VRd. Fig.
6.55 fc as Page 617 Table of contents .ct = 0.2. since the capacity of such a member results to be independent of the slenderness ratio a/d. 6. The dotted plane is valid for a maximum stress 0. Consequently. a/d ratio’s and support widths.2. chapter 6. Loads near to supports In 6. 6. 2001 version.According to the formulations for the strut and tie model the capacity of the concrete struts only depends on the strength of the concrete.DI cp fctd (6. Fig.18: .17) S is obtained. Furthermore short members are prone to significant size effects. It was shown [Lehwalter and Walraven.19.12 k (100 Ul fck)1/3 (2d/x) bw d.2. since it might be argued that loads near to supports may be treated with the rules given in EC2. is extended with a factor (2d/x) in order to cope with the increased shear capacity in the case of loads applied near to supports.2. Lehwalter carried out tests on short members with various sizes. see e. . It is seen furthermore that the limit 0.6 fc. 1994). which is known to have a strong influence. and compared the equivalent maximum stress in the concrete struts.5d < x < 2d the shear capacity may be increased to VRd.g. that the size effect in short members is the same for short and slender members. there are many arguments in favour of the formulation according to Eq. It is seen that for lower a/d ratio’s the capacity is considerably higher than the one obtained with the strut and tie model. at a distance 0. so that here also the factor k = 1 + ¥(200/d) applies.20.19. 6.18) This may need some explanation. the maximum capacity is a function of the concrete strength and the width of the support area. (6. However. fig. According to this formulation.2 (5) the equation (5) or in prENV 199211:2001 the Equation 6.3.a. Bearing capacity of short member according to strut and tie model with defined maximum concrete stress in the struts It can easily be seen that this is a very simplified representation of reality.5 “Design of discontinuity regions with strut and tie models”. Figure 6.
but not 95% of the points lie above Vtest/Vcalc = 1 for 2d/x. 1989) Figure 6. For continuous beams with concentrated loads even (3d/a) gives safe results. A more accurate formulation than the strut and tie model is therefore useful in those cases. like in the case of corbels and pile caps. Without a provision like the one given in Eq.5 5(P) for struts with transverse tension is appropriate for a/d < 2. members with depths until 1 m and a support width up to about 0. it is important to reduce the size as much as possible. with Vcalc computed assuming values of 2d/x and 2. For simply supported beams subjected to distributed loading only (2d/a) gives safe results. so that the governing shear load is small. For a number of practical members.5d/a) is appropriate. Lehwalter. Matthey and Watstein (1963). Regan (1971) and Rogowski and MacGregor (1983) b.21. De Cossio and Siess (1960). with a slab of about 1 meter and wall distances of about 5 m. Clark (1951).Another case is shown in Fig.Guide to EC2 Section 6 defined in 6. Lehwalter (198 ). Küng. The results which are on the unsafe side with 2d/x need not to be of too much concern as Krefeld and Thurston’s work includes many beams with low concrete strengths. 6. It is a part of a foundation caisson in the Storebaelt bridge. concluded that: a. Figure 6. or a uniformly distributed load and a concentrated load) can be handled. An important question is whether the multiplication factor should be (3d/a. Rüsch.21. Most.22 on the basis of tests by Baldwin and Viest (1958). Haugli. also combinations of loads (like two concentrated loads.20. Leonhardt and Walther (1961). 2.23 is based on tests by Bernaert and Siess (1956).5d/a or 2d/). The diagram in Fig. Morrow and Viest (1957). A substantial part of the counterpressure of the soil is transmitted directly to the walls. Page 618 Table of contents . . unnecessary shear reinforcement would be required. For simply supported beams subjected to concentrated loads a factor (2. This is confirmed in Fig. The figure shows graphs of Vtest/Vcalc plotted against l/d. Foundation slab in Storebaelt caisson By introducing the distance x and determining the shear capacity in every cross section. on the basis of the analysis of many experiments.25d.4.0. but less than half do so with 2.5d/x. c. Regan [1998].5d/x. 10. Maximum stress in concrete struts as calculated on the basis of test results (Walraven. 6. 6. Mayer (1962) and Krefeld and Thurston (1966).
Exceeding the tensile strength of concrete in regions uncracked in bending as described by expression (6. Results of tests on simply supported beams without shear reinforcement subjected to distributed loads (Regan. 3. 4. Exceeding the shear resistance given by expression (6. Page 619 Table of contents .Guide to EC2 Section 6 Figure 6. Results of tests on simply supported beams without shear reinforcement subjected to concentrated loads (Regan.23.2a) in presence of bending moments greater than the cracking bending moment. when the acting shear corresponding to the cracking bending moment exceeds the bearing capacity calculated with expression (6. Prestressed members without shear reinforcement Four failure modes may be envisaged under combination of shear and bending within the extremity region for such elements: 1. Snap back failure at cracking bending moment outside the transmission length.2.2a). The designer should verify all the four failure mechanisms described above with particular care to the snap back behavior.4). Anchorage loss due to bending cracks within transmission length.4.2. 2. 1998) Figure 6. 1998) 6.22.
In ENV 199211. rotation can only continue until crushing of the concrete occurs. This method. activate more stirrups for the transmission of the shear force and.4. Fig. resulting in strut inclinations smaller than 45°. If the shear reinforcement yields. Ultimate capacity Vu. with strut inclinations smaller than 45° may be assumed for design.25. taking account of the fact that the beam web.2. A result of the smaller strut inclination is.9d s stirrup distance fyw yield stress of stirrups inclination of concrete struts Asw crosssectional area of one stirrup Q effectiveness factor.1 Non prestressed members In ENV 199211 the Standard Method and the Variable Inclination Method were offered as design alternatives. is from a physical point of view unsatisfactory.24. If the strut inclination is and the (vertical) shear reinforcement yields. in chapter 4. so that an appropriate upper limit to the shear capacity has to be defined. a shear force Asw z fyw cot Vu. that a redistribution of forces occurs in the webs of shear reinforced concrete beams. by rotation of the compression struts to a lower inclination. The choice of the lowest value mostly leads to the most economic design. Consequently. the stress in the concrete struts increases.3. b z fc1 Vu. giving a transparent view of the flow of forces in the structure.2.2 = w (6.5 which means that may be chosen between 21. as such. 6.2°. is not as well suited to resist the inclined compression as cylinders used to determine fc.19) s Is transmitted. The Standard Method.25b).4. however. the truss can. the corresponding shear force is (Figure 6.24. that the stresses in the concrete struts are larger.3 = (6. Due to strut rotation.Guide to EC2 Section 6 6. Fig.3 Members requiring shear reinforcement.8° and 68. This reality is.2. 6. 6. a larger number of stirrups is activated.25a.3 for yielding stirrups b.3. extend the zone of failure. because the “concrete term” is purely empirical and hides the physical reality.3 Members requiring design shear reinforcement C6. Fig. and is known as the “variable strut inclination” method. This approach is not only attractive because of its agreement with the physical reality. the designer is allowed to choose the strut inclination between 0.2. In this case the compression struts are supposed to rotate from an initial value of 45* to a lower value of about 22°. but also because it is a simple equilibrium method. Figure 6. Redistribution of forces in a shearloaded web by strut rotation Because of a smaller strut inclination. 6.4 < cot T < 2. For an ultimate compression stress fc1 in the concrete struts. a. and the shear capacity is increased. Ultimate capacity Vu.20) cot + tan where fc1 = Qfc In those equations bw web width z inner lever arm # 0. in spite of its acceptable accuracy. which is transversally in tension.2 for crushing of concrete struts Page 620 Table of contents .
21) is used (CEBFIP Model Code 1990). By equalling 6.6 (1 – fck/250) (6. Figure 6. Fig.20 the maximum possible shear force and the corresponding inclination are found.26.3 on the strut inclination The expression for the ultimate nominal shear strength vu and the corresponding value of are then Qu (6.22) \ 1.2 and Vu. 19 shows the development of Vu. Dependence of Vu.Guide to EC2 Section 6 For Q generally the expression Q = 0.2 and Vu.19 and 6.\ .3 for decreasing .
\ Where Vu Qu = (6. fc1 And \ (6.9d .23) tan = 1.24) b w 0.
6. Figure 6.\ coordinate system. By rotating down to a lower inclination the beam activates more stirrups to carry the load. When the crushing strength of the struts is reached the beam fails in shear. The behaviour is elastic: the (constant) principal strain direction depends on the “stiffness ratio'’ in the cracked state. with cutoff for cot = 2.25) with fc1 = Qfc and w is the shear reinforcement ratio according to w = Asw/bws. Graphical representation of Eq. 6. Eq.28. For \ > 0.22 represents a circle in a Vu/fc1 . runs from 0 to 45°.27. Line 2 At the formation of inclined shear cracks the principal strain direction decreases Line 3 After having reached the stabilized inclined crack pattern a new type of equilibrium is obtained. \= sw yw f fc1 (6. In graphical terms this is shown as the linear cutoff in Fig. 16. 6. 6. Page 621 Table of contents . Fig. the value of is constant and equal to 45°. However. The diagram shows lines numbered from 1 to 4.5. Line 1 In the beginning of shear loading the beams is uncracked in shear so that the principal strain direction is 45°. Line 4 When the stirrups start yielding the web searches for a new state of equilibrium.5. In the mean time the compressive stress in the concrete increases.5 Measurements of the deformation of the web in shear loaded Ibeams show typically a behaviour as shown in Fig.27.5 is a limit. It is found that when \ runs from 0 to 0.27. it was assumed that cot = 2.
1999) Fig. Rotation of the concrete struts as measured on the web of beams with shear reinforcement.Guide to EC2 Section 6 Figure 6. Hamadi and Regan (1980). 6. Regan and RezaiJorabi (1987). 1995 and 1999) Fig.19 and Eq.28. Fig. 6. a Strut rotation as measured in beams with normal strength concrete (Walraven. Figure 6. Levi and Marro (1993) and Walraven (1999). schematically represented (Walraven.30 shows a verification of the combination of Eq. 6. M = medium and H = High).29a shows the strut rotation as measured in three beams with different shear reinforcement ratio’s (L = low. Leonhardt and Walther (1961).Strut rotation as measured in beams made of high strength concrete (Walraven.20 with the limit cot = 2. 1995) b.29. It can be seen that for M and H the beams fail before the stirrups have yielded. Placas and Regan (1971). 6. Moayer and Regan (1971). Muhidin and Regan (1977). Figure 6. Non prestressed beams with vertical stirrups – relationship between shear strength and stirrup reinforcement Page 622 Table of contents . Kahn and Regan (1971). 6.30.5 with test results from Sörensen (1974).29b shows a similar diagram for a series of high strength concrete beams (fc § 90 N/mm2).
Clause 4. at no section in any element should the design shear force exceed VRd. Leonhardt. Aparicio and Calavera (2000). Where the member is subjected to an applied axial compression.2. VRd2 should be in accordance with the following equation: VRd.3. Prestressed members with shear reinforcement If the same rules are applied to prestressed members with shear reinforcement.67 VRd. Görtz and Hegger (1999).eff/fcd) < VRd. Reduction of maximum shear capacity by axial compressive stress according to ENV 199211.3. The following statement is found: “In the absence of more rigorous analysis.Guide to EC2 Section 6 6. Koch and Rostasy (1974). it can be seen that there is apparently an increase of both safety and scatter. 1987).2 (6.26) Fig. Fig.2 (4) Page 623 Table of contents .32 shows the dependence of the upper limit for the shear capacity on the level of prestressing. Moayer and Regan (1974). The test used for this figure are from Hanson and Hulsbos (1964).30. Experimental results of shear tests on prestressed beams with shear reinforcement. described in (Walraven.32.2 (1 cp. like in ENV 199211 and MC’90. Figure 6.3.red = 1.i is partially neutral and partially negative. 6.31. In 4.2. 6. Lyngberg (1976).2. Fiugure 6. Many of the test results are collected in a databank. Levi and Marro (1993). in comparison with the calculated results according to the variable strut inclination method (no special web crushing criterion for prestressed concrete) In ENV199211 the effect of prestressing on the upper limit of the shear capacity VRd.2.2 (4). Bennett and Debaiky (1974).2.
34. with 0.29) However.33. 6.0 cp.Guide to EC2 Section 6 In his “Commentaries on Shear and Torsion”.4fcd < cp < fcd Figure 6. Fig.5 (6. 6.for large compression.28) . It appears that the safety margin and the scatter are reduced (the details of the calculation are found in Appendix 1).cp/fc) for 0.for small compression.34. have ratio’s of cp/fc lower than 0. 6. is Dc = (1 + cp/fc) for 0 < cp/fc < 0. with 0 < cp < 0. that the expression is not sufficiently conservative. 6.5) with a factor Dc = 1 + 2. according to Nielsen (1990) Dc = {1. for the region of low compressive stresses this expression gives about the same results as Eq. taking into account the influence of prestressing.5 (6. 6.67 cp/fctm) (6.29. Comparison of results of shear tests with variable inclination truss analogy with effectivity factor Q = (0. however. Figure 6.199211 (without influence of prestressing) and proposals by Nielsen (1990) and Fouré (2000) for fck Page 624 Table of contents .29) with original formulation in ENV.0fc In Fig.7 – fck/200 > 0.eff / fc < 0.31 are evaluated using Eq. 6.27) A comparison with 93 tests results shows. Another proposal for taking the influence of prestressing into account in the Qvalue was given by Fouré (2000): . 6. He proposed to multiply Q (which was then equal to Q = 0.25 for 0.33.5 (1 . used in the comparison. It should furthermore be noted that all test results.25fc < cp < 0.4 and that the multiplication factor is obviously applicable only for this region.35 the same data as used in Fig.2 (1  cp/fcd) (1 + cp/fcd)} 0.25 fc Dc = 1.27 A more moderate expression.7 – fc/200)(1 + cp/fc).eff / fc with cp.5fc < cp < 1. to serve as a safe lower bound over the whole region of test results. Nielsen (1990) states that prestressing has a positive influence on the shear capacity of beams with shear reinforcement. The new proposal is compared with the other ones in Fig.5fc Dc = 2.4fcd Dc = (1 – 0. Comparison of new proposal (Eq.
36. 6.3. in comparison with the calculated results according to the variable strut inclination method. (Asin. 2000).29 6.35.2. Combination of truss and strut and tie Measurements on shear reinforcement showed that the stirrups just adjacent to the load. In the Standard Method.and support area do not reach the yield stress.75 area between load and support. 6.3 Torsion See example 6.Guide to EC2 Section 6 Figure 6.7 Page 625 Table of contents . with extension according to Eq. Experimental results of shear tests on prestressed beams with shear reinforcement. Members reinforced in shear with loads near to supports Similar to members without shear reinforcement.ct + Asw fywd sinD The transmission of forces occurs according to Figure6.36. the Variable Inclination Method not containing a concrete term. as formulated in ENV 199211 this was taken into account by multiplying the “concrete term”. in members with shear reinforcement the load bearing capacity is increased for loads near to supports. (6. The formulation for this case is then VRd = VRd.30) Figure 6. it was introduced here for x/d < 2. However. Therefore the shear reinforcement is considered to be effective only within the central 0.6 and 6. with a factor (2d/x).3.
25 fctk /Jc k size factor = 1.6.37. The design punching shear capacity was VRd1 = vRd1·u (6. where also Eq. Altogether 112 test results have been considered. First it makes the limiting shear stress much more uniform for different column sizes.32b is applied. taken from [212]. 6. 6. the control perimeter is moved from a distance 1.4 Punching C6. when the design punching shear capacity is given by VRdc = vRdc .0 d in mm l = ¥( lx . Table 6. to a distance of 2d. which leads to unconservative results for higher concrete strengths.32b) (6.31a) Where u is the length of the critical perimeter.02 fck = characteristic cylinder strength of concrete In the new definition. whereas 32 refer to tests on high strength concrete specimens with concrete cylinder strengths ranging from 60 to 120 MPa.12 k (100 lfck )1/3 with k = size factor = 1+ ¥(200/d) 2.9. 78 of those results refer to tests on specimens with cylinder strength ranging from 15 to 60 MPa. (6. There are two reasons to adopt the distance 2d.12 in Eq.4.31b) where WRd basic shear strength = 0. Therefore at first an evaluation is carried out in order to verify this value. taken at a 2d distance from the loaded area d = mean effective slab depth = (dx + dy)/2 and where the design punching shear stress for nonprestressed slabs value is vRdc = 0.ud where u = length of the critical perimeter.9. the design shear resistance per unit length vRd1 followed from vRd1 = WRd k (1.2 + 40 l)d (6.5d from the column.4.0 d effective depth of slab = (dx + dy)/2 l flexural reinforcement ratio = ( x + y)/2 It was already shown in 1980 [1].5d from the loaded area. that in the derivation of this equation an error was committed.Guide to EC2 Section 6 6. which was defined to located at (1.5d) from the edge of the column. This enables a good evaluation of the validity of the punching shear formula (Eq.37. The tests cover the interval of individual parameters given in Table 6. Basic control perimeters around loaded areas Often questions are raised with regard to the coefficient 0.32a) Figure6. Therefore it was decided adopt the formulation for the punching shear capacity given in Model Code 1990. 6.d [m] > 1. taken at a distance 1.32a and 6.32b) for higher concrete strengths. Punching shear capacity of nonprestressed slabs without punching reinforcement In ENV 199211 a nominal shear stress was defined as the load divided by the product the slab's effective depth and the length of the control perimeter. 6. Second now for punching the same formulation can be used as for normal shear in members without shear reinforcement.1. according to MC’90. Range of parameters in tests used for evaluation Page 626 Table of contents .32b. Basic equation for symmetrical punching at interior columns 6. ly) 0. see Fig.
5 as well for the derivation of the punching shear capacity would be inappropriate.32b of 0. Figure6. Those tests therefore have to be regarded with some reservation. Figure 6. a reliable design equation can be derived from test results with the general formulation BRd = BR (1 . assuming a safety factor of 1.38.8 and BR = 0. The tests of Base fall somewhat outside of the scope probably due to the very small maximum particle diameter (Dmax = 4. 6.8 in the case of one dominating parameter BR target safety index.6. and a coefficient of variation v = 0. 6.116 Page 627 Table of contents .39 shows the frequencies of the relative punching shear capacities (ratio experimental to calculated values) for the data considered.130 a value for the design coefficient in Eq. 6. Therefore a more sophisticated approach was necessary in order to unambiguously derive a design equation with the required level of reliability. although the variation of the concrete strength (for which a material safety factor 1.0247 = 0. normally taken as 0.0.12 in the equation of the design punching shear stress (Eq. 6.0247. so that simply applying a material safety factor 1. According to the level 2 method.191 . 6.191. taken 3. a mean value of 0.32 Fig. as described in ECl Basis of Design is suitable.8 coefficient of variation BR with BR = 0.32b ). The same method was applied by Konig and Fischer for investigating the reliability of existing formulations for the shear capacity of members without shear reinforcement [14].10 in stead of 0. BR = 0.39.BR BR) (6.164 0.5 this would result in a coefficient of 0. For such a case the classical "level 2 method". Frequencies of relative punching shear carrying capacity Assuming a normal distribution.38 shows a diagram in which the values vexp /vcalc are shown as a function of the concrete strength. Strictly speaking this would mean a characteristic lower bound value of 0. However.13.32b). Punching strength of slabs without shear reinforcement: comparison of test results v/s eq.15. The way how to deal with this method has been described and illustrated by Taerwe [13].5 applies) is the dominating factor with regard to the scatter of results in punching shear tests. the punching shear capacity is not linearly proportional to the concrete compressive strength (the fck value has an exponent 1/3 in Eq.Guide to EC2 Section 6 Fig.2 mm).191 is obtained with a standard deviation of = 0.33) where BRd design value mean value of test results BR sensitivity factor for BR.
13 for the punching tests this would mean an increase of the coefficient 0. according to the relation vu > C fctd.b is correct. This would then result in a coefficient 0.116 of about 6. Eq. Page 628 Table of contents . see also the report for shear. This means a coefficient of variation for a concrete C25 of v = 0. Pralong.0. It can therefore be concluded that equation 6.4. because the contribution of prestressing to the punching resistance depends also decisively on the definition of the control perimeter.15 and for a concrete C90 of v = 0.0. Taking into account this lower li the design equation for nonprestressed slabs should therefore be: VRd.c = 0.37) Adding this contribution to the punching shear capacity of a similar slab without shear capacity.05. In combination with BR = 0.20) vRd.c (prENV 199211:2001 6.8 % for C25 and 3.120 for a concrete class C90.c = (0.15 cp to the ultimate shear stress vRd. Fig 6.35fctd (6. Nölting [44].32. 6 with test results 6. A disadvantage of Eq. 6. Gerber & Bums [41].18/Jc) k (100 l fck)1/3 > 0. Verification of Eq. that may for instance occur in prestressed slabs.c = (0.8%) too high.8 and = 3. This could give unrealistically low punching shear capacities for low reinforcement ratio's.34) is given.Guide to EC2 Section 6 was obtained. Therefore a lower bound was added.a. where fctd = fctk /Jc.58 . 3. solely depending on the concrete tensile strength. It is logic that also with regard to punching the effect of prestressing will be positive. giving values which are only slightly (2.4.40. Furthermore designers like to have a simple lower bound formulation for a fist check.58 and the standard variation is s = 0.35 fctk /Jc .8 the design value should be (1.2. seems to be quite reasonable. In the Model Code the relation fck = fcm .32a. 91). is acceptable as a design equation.3 Punching shear resistance of labs with punching shear reinforcement In ENV 199211.42.8 (MPa) (6. It should however be noted that the collection of tests does not contain slabs with crosssectional depths larger than 275 mm.124 for a concrete class C25 to 0.38 with the equation vu = C fctk.34. 6. Stahlton [42].b is that they go to 0 if U goes to 0.8 . used as a lower limit for shear as well.33.c. It is however not expected that the same term as for shear can be used.36. Actually this means that Eq.972 vRd.57 applied and for high strength concrete (>C50/60) a value equal to C= 0.35.20.2.2 Punching shear resistance of prestressed slabs without shear reinforcement. The value 0.6 % for C90 (see [14. Brändli. 6.8 . 6.18/Jc) k (100 l fck)1/3 .08 cp (6. 6. p. 0. Thürlimann [43] and Kordina. with = 0.36) The mean value of vexp/ vRd. the mean concrete cylinder compressive strength has been used.08 cp > 0.0. In order to cope with larger slab depths used in practice the coefficient C 5% should therefore be further reduced.2a).35) 6. A comparison of the test results with the design equation: vRd. it was assumed that the contribution of punching shear reinforcement to the total shear capacity can be accounted for by Asw fyd sin D (6. For shear loaded members the influence of a normal compression force is taken into account a separate contribution of 0. According to Eq. In order to find the contribution prestressing a selection of test results has been made: Andersson [40]. In this derivation. it was found that for normal strength concrete (<C50/60) a 5% lower value C =0.c is 1. Evaluating the same results as shown in Fig. whereas in the code expression the 5%lower value fck is used. however.
Figure 6.75 vRd. according to fywd. 6.c fywd. One group of researchers.4 vRd1. In the prENV 199211:2001./Nölting [18] propose efficiency factors for the contribution of the shear reinforcement ranging from 0. Yitzakhi [25] and Regan [24] with the formulation Vu = 0. but with effective design strength of the shear reinforcement which depends on the slab depth. like Elstner/Hognestadt [19].41.25d < fywd (Mpa). Fig. This upper limit will. This means that VRd.eff Asw (6. Punching capacities related to resistance of the shear reinforcement (Regan [24]) The increase of the punching shear capacity by shear reinforcement is limited to a bound.c ud + Asw fywd. However. Pranz [6]. however.25. propose to use the summation principle. Herzog [16] Petcu [17]. such as Moe [15]. Therefore a Page 629 Table of contents . 6. so that the failure load is increased. in order to account for the anchorage efficiency. which can be more efficient than stirrups as shear reinforcement. In literature two types of proposals are distinguished in order to cope with this phenomenon. 6. 6.eff = 250 + 0. In a redraft in 1991 this value was increased to 1. A further argument against the taking into account in full of the reinforcement according to Eq.cs = 0.31b. One can also say that due to the vertical movement of the punching cone.6 vRd1. Kordina.max = 1. Others. A first important question is if the summation principle of a concrete component and a punching reinforcement component is valid anyhow.80 down to even 0. In the version of EC2 of 1988 the upper bound was formulated as vRd.Guide to EC2 Section 6 according to Eq. higher values of the upper punching capacity than 1.39) according to Eq. where vRd1 follows from Eq. shear reinforcement within the perimeter considered.75Vc + Vs (6. In the descriptive model of Kinnunen and Nylander [15] for slabs without punching reinforcement the tangential compressive strain at the bottom face of the slab is the design criterion. 23]. using shear heads and shear studs.41 shows a comparison of test results (punching failures) from Gomcs [22. where normal punching shear reinforcement is used.eff sin where vRd. A punching shear reinforcement would then only be able to limit the rotation of the kinematic punching shear mechanism and as such reduce the compressive strain in the critical area.and Regan [20]. however with a reduced concrete contribution (efficiency factors ranging from 0. concrete component has already a reduced value at the moment that yielding of the shear reinforcement has been reached.34 is that it is hard to find adequately anchored punching shear reinforcement at both sides of a critical crack Therefore the punching shear reinforcement is not yet yielding when the concrete contribution is at its maximum. would gives the total punching shear resistance. hardly be reached in practical situations.38) where Vc is the concrete contribution (punching shear capacity of similar slab without shear reinforcement) and Vs is the contribution of the yielding steel.6vRd1 can be reached.6 to 0. 6.8). the formulation according to MC’90 has been chosen.31.32b inclination of the shear reinforcement design strength of shear reinforcement.
and Q is equal to (6.43) where VEd design value of the punching shear factor taking account of the expected effect of eccentricity. 6.42 Figure 6. It should be verified that no punching failure occurs outside the outermost layer of shear reinforcement.42) where vEd is the design calculated of the ultimate shear stress on the perimeter un. since steel closer than this will not be well anchored in the compression zone if intersected by cracks at lower inclinations (Regan [24]). Finally the shear reinforcement (with the same cross section per unit area) is extended to a distance 1. 6.d) (6. The design punching shear stress vEd was formulated as vEd = ( VEd)/(u.08 cp (6.43.41) The distance from the column to the inner shear reinforcement should not be larger than 0.3d. Then the perimeter is determined for which vEd < vRd.12 k (100 l fck)1/3 . see fig. At first at the perimeter with distance 2d from the loaded area the punching capacity is checked and the eventual shear reinforcement is calculated.5d from the outer perimeter.c.4.42. according to VRd.44) where W1 is a function of the control perimeter u1: W1 ³0u1 e d A The property W1 corresponds to a type of “plastic” distribution of the shear stresses as illustrated in Page 630 Table of contents . Approximate values for eccentricity factor E in new draft Complementary to this simplified approach the more accurate method.43 the values for the eccentricity factor are given. This method takes the effect of an unbalanced moment into account with the formulation: VEd = VEd/(u1d) + (KMEd)/(W1d) (6. 6. No further indication on what this “more rigorous analysis” means was given.5d.5Q fcd u0 d where u0 is the length of the column periphery. The distance between the layers of shear reinforcement in radial direction should not be larger than 0.75d.Guide to EC2 Section 6 modified upper limit has been defined according to MC’90. Therefore an additional perimeter un is defined at a distance of 1.40) Q = 0. It should be shown that here vEd < 0. In Fig. given in MC’90 has been adopted in prENV 199211:2001.0. Figure 6.5d from the outermost shear reinforcement. “in the absence of a more rigorous analysis”. Basic equation for eccentric punching In the draft of 1988 only a very general approach as in combination with a bending moment. nor should it be less than 0.3. Control perimeter un at interior column Practically the design of the shear reinforcement is quite simple.60 (1fck /250) (6.max = 0.
45 quite well. 11 with test on interior column – slab connections The method can be as well applied to edge and corner columns. An analysis by Mast [34. Fig. Reduced perimeters for assumed uniform shear for edge and corner columns Page 631 Table of contents . 6.36]. in those cases the ultimate punching shear stress can as well be calculated in a simplified way assuming uniform shear on the reduced perimeter shown in Fig. Shear distribution due to an unbalanced moment at a slab internal column connection K is a factor taking into consideration that a bending moment in the slab. The diagram is a combination of two figures. However. Comparison of Eq.44. Figure 6. For a rectangular column W1 follows from W1= c12 + c1 c12 + 4c2d + 17.41 in comparison with tests results. is only sustained by bending in the column but also by bending and torsion in the slab itself.40.8d2 + 2Sdc1 (6.46.35] on the basis of an elastic analysis of the distribution of shear stresses in a slab in the vicinity of a column showed that those stresses approach the distribution shown in Fig. 6.45.45 shows Eq.Guide to EC2 Section 6 Fig. 6. 6. from Regan [24.46. Fig 6.6. 6.45) Figure 6. With this approach the shear capacity depends on the column size and the value of the unbalanced moment and is therefore much more accurate than from Eq. K follows from the table below For round columns c1/c2 = 1 so K = 0. 6.44.
8 to 6. It turns out that the ultimate column load is a function of the based to column width l/c but is independent of the ratio c/d. which means indeed that the inclination of the punching cone is much steeper than in suspended slabs.47 the design column load VEd can immediately be determined as a function of the ratio l/c. Figure 6. This was not regarded in the EC2 version of 1988. It can be seen that in the utmost number of cases the value of acrit is smaller than 2d.60 (1 .Guide to EC2 Section 6 6. 6. the inclination of the punching cone in column bases may well be steeper than in suspended slabs.e.48) where a distance from the periphery of the control perimeter considered Q 0. For many combinations of base width to column width l/c and column width to effective slab depth c/d the critical ratio acrit /d has been determined.6 Anchorages and laps 6. 6.46) where VEd column load VEd the upward force within the control perimeter considered i.1 Punching shear in column bases The most important difference between punching of a slab around a column and punching of column base supporting a column is the presence of a significant counter pressure from the soil.5 Design with strut and tie models 6. For concentric loading the design punching force is VEd. The nominal ultimate shear stress at the perimeter is vRdc = 0.8 Fatigue See example n.47. which gives rise to uncertainty with regard to the critical perimeter to be checked. A second difference is that the distance from columns to the edges of their bases are commonly much smaller than those to sections of radial contra flexure in suspended slabs [38].47 shows a result of a parameter study.VEd (6.red/ud (6. The results are shown in Fig. 6.12 k(100 fck )1/3 2d/a < 0. In the lower diagram of fig. upward pressure from soil minus self weight of base vEd = VEd.3.37). in which the position of the control perimeter is treated as a variable and the unit punching resistance.15 No comments No comments Page 632 Table of contents .47) where u is the control perimeter taking a value a < 2d instead of 2d into account (see also fig. for which the lowest column load is obtained.red = VEd . 6. i.49) Fig. with the inclination of the failure surface. The corresponding value of the critical control perimeter acrit /d is read in the upper diagram. 6.5Qfcd (6.4.47. The CE Model Code 1990 gives an alternative. Due to the influence of the vertical soil pressure. Shear capacity of column bases 6. carried out with the previous equations.14 See example n.7 Partially loaded areas 6.fck /250) (6.e. 6. is taken to vary with the distance from the column to the control perimeter.
No.A. (1951).M. June 1964. Koch. (1961). pp. pp. J (1999). (1966).. CEB Bulletin 224. G. “Schubversuche an Spannbetonträgern”.. B. ACI Journal. De Cossio.2 (Ultimate Limit State – Shear”. and Siess. ACIJournal. 53 No. 184209. (1958). Rostsy. Darmstadt University of Technology . Asin. Nov. (1974). No. Vol.P. and Thurston. October 1951. Page 633 Table of contents . 11/1974 Baldwin. Görtz. Deutscher Ausschuss für Stahlbeton. (1993). ACIJournal. Chen Ganwei (1988). ACI Journal Sept. Küng. F.. V..”High strength steel as shear reinforcement in prestressed concrete beams”. University of Denmark.(1971). S. Collins.D.. (1964). S. Koch. December 1980. “Effect of axial compression in shear strength of reinforced concrete frame members”. “The bearing capacity of concrete struts in strut and tie models exemplified for the case of short beams”. 695735. 451476. König. . “The Stuttgart shear tests 1961”. 293305. (1980).(2000) “Testing strut compression shear failure in beams”. R. 1960.. Sandefjord. Jan 1957. pp. 49100.24 June. pp. 1958.J. Levi.J. Symposium Utilization of High Strength/ High performance Concrete.Elstner.. F. Bennett. (1960). F. PhDThesis. 145156. I.N. ACI Special Publication 42. “Plastic Analysis of Shear in Beams.P. Series of papers in Beton . I.E. “Behavior of normal and lightweight aggregate beams with shear cracks”.. Norway. A. “Studies of the shear and diagonal tension strength of simply supported reinforced concrete beams”. Luxemborg. pp.E.Y.. pp. Marro. (2000). ACIJournal. “Design of concrete structures with regard to shear forces”. pp. Fouré. “The behaviour of reinforced concrete deep beams”.. Lillehammer 20. S. 1989.M. Beton. R. Heft 227. Hulsbos. Leonhardt. P. (1966). 1988. R. Berlin. R. 2024 June 1999. 63. 1974.. (1957). Bernaert. “Strength of shear in reinforced concrete beams under uniform loading”. and Regan.. 30. M. “Behavior and strength in shear of beams and frames without web reinforcement”. University of Illinois. Balasoorya.Oct. ACIJournal 64. May 2000 (unpublished). Bhide. Calavera. E.. Clark. proceedings.und Stahlbetonbau. S.. Vol. Fischer. (1974).L. “Laboratory investigation of rigid frame failure”. “Proposal for rewriting item 6.. PhDThesis. Proceedings. pp. O. pp. “Model Uncertainties concerning Design Equations for the Shear Capacity of Concrete Members without Shear Reinforcement”. Betonstahl in Entwicklung. pp. pp. B.2. Hamadi. ”Basic Facts Concerning Shear Failure”. C. (1966). 637668. 126. ACI Journal.. No. The Structural Engineer. No.”Ultimate shear tests on prestressed concrete Ibeams under concentrated and uniform loading”.und Stahlbetonbau. Bennett. (1973). O. June 1956. Vol. Serie R. . 1.Guide to EC2 Section 6 REFERENCES Shear Aparicio. pp.D. Aster. J. del Pozo. Viest. R. Germany (in German). Krefeld. Department of Structural Engineering. Heft 33.. The Netherlands. 58B No.B. (1988). Walther. Deep.W . July 1995.W. J. No. P. G. “Ein Beitrag zur Schubsicherung im Stahlbetonbau”. 675 – 692..W. F. Vol. Siess. Losberg. (1978). A.Debaiky. 6. Leonhardt. 1978. Delft University of Technology.1528 Hedman. E.. “Shear tests on HSC prestressed beams – proposals of new interpretative models”. Kani. M. 7179.A. “Shear Capacity of Prestressed Concrete Beams of HSC Elements”. pp. ACI Journal Vol.P.. Feb. Vol. Y. 312321. “Diagonal tension in reinforced concrete beams”.M. April 1966. 570581. 4. W. Hegger. F. TorEsteg Steel Corporation.4 in section 6. Lehwalter. (1995). March 1971. Hanson. Conference on High Strength Concrete 1993. 231248. 237. 1. PCIJournal. 4.”Shear strength of prestressed beams with thin webs failing in inclined compression”. Beams and Corbels. C. Technical. C. 635654. J. CEB Bulletin d‘Information No..C. R. N. Hognestadt.J. Polytechnic University of Barcelona. 4. pp. “Schubtragfähigkeit dicker Stahlbetonplatten”. (2000). H. 2. 32. Note to Project Team for EC2.(1989)’’Influence of Axial Tension on the Shear Capacity of Reinforced Concrete Members”. J.(1956).W.
Regan. P. (1978). “The shear resistance of reinforced concrete Ibeams”. D. 11. Walraven. shear and torsion”. (1990). Proceedings Vol.”Enhancement of shear resistance in short shear spans of reinforced concrete”. pp. M.. 1. Watstein. Vol 1. 8187 (in German). Sandefjord. “Shear tests on 12 reinforced concrete Tbeams”.44/95. Fibre reinforced materials – design and engineering applications. H.E. P.E..C.S. report R60. Studi e Ricerche. Politecnico di Milano.E. Sandefjord. 1963.. (1977). 1978. N. pp. pp. Walraven. (1983). 2024 June 1999. P. Structural research Laboratory. 305321. (2000). 2000. Muhidin. 1994. 144 pages. RezaJorabi. “Toward a consistent treatment of model uncertainties in reliability formats for concrete structures”. Technical University of Denmark. . J.E. (1987). Vol. 1998 (14 pages).Structural Journal. Norway.”Shear strength of deep reinforced concrete continuous beams”. (1971). Proceedings of a conference.C. ACI Journal. University of Alberta.M. (1962). an evaluation of UK recommendations and particularly of BD. Morrow. Chew.G. J. 13. Heft 145 (in German). N. Betonund Stahlbetonbau.. J. D. Mayer.B. 2. pp. Moayer. Shear capacity of high strength concrete beams with shear reinforcement”. Feb. pp. 1974.E. 135 . AlZubi. Proceedings Symposium on Lightweight Aggregate Concrete. pp. “Aspects of diagonal tension in reinforced concrete”. “The influence of member depth on the shear strength of normal and lightweight concrete beams without shear reinforcement”.. 141223. Technical Note 45.Oct. Sept. “Ultimate Limit State Principles: Basic design for moment.. “The bearing capacity of concrete struts in strut and tie models exemplified for the case of short beams”. MacGregor. 2001. “Shear in prestressed concrete”. No. Regan. (1963).G.W. “Shear in reinforced concrete – an experimental study”. P. London 1977. J. “Shear strength of various shapes of concrete beams without shear reinforcement”.. pp. (1989). Delft University of Technology.. 60. in fib Text Book on Structural Concrete. Regan.E. Regan. Structural Engineering Report. (1993). “Safety and Performance Concepts”.E. “Shear strength of beams without web reinforcement containing deformed bars of different yield strengths”. April ’89. Eurocode 2 editorial Group – 1st draft – October 1990. CIRIA. (1995). F. Rüsch. University of Westminster.C. 180. (1987). Vol 19. Symposium on High Strength Concrete. “Commentaries on Shear and Torsion”. No. Regan. 1983.. pp. and Regan. Journal of fib. P. 1976. P. H. Lehwalther. pp. Page 634 Table of contents . Nov. Viest. 1987. 183207. No. Haugli. 91104 Walraven.. Lyngberg. 585593. Structural Concrete. pp...A.C.”Shear capacity of lightweight concrete beams with shear reinfocement”. April 1976. J. J. P. 9. Stroband. N. Y. 833869. (1999). L. N. Sörensen. pp. ACI Special Publication 42. Eurocode: Basis of Structural Design. Mathey.C. Heft 4. J. H. (1994). Vol. and Regan. 1.R. ACI Journal Vol. (1999).P. (1974). “Chopped steel fibres as shear reinforcement in concrete beams”. (1998). Lehwalther. London. Stage 34. Institution of Civil Engineers. Norway. CEB Bulletin d’Ínformation 219.. (1976). Walraven. P. B. Rogowski. “Shear tests on rectangular reinforced concrete beams with uniformly distributed loads”.. 183221.. Sept.M.. London. 91. C. Walraven. 119132.R.C. H. J. (1974) ”Shear strength of prestressed and reinforced concrete Tbeams”. “Size effects in short beams loaded in shear”. and Regan.149. 2. Ultimate shear resistance of partially prestressed concrete Ibeams”. No. Vol. AC_ .. CEBBulletin d’Information Nr. 214221. No. pp. (1957). Stevin Report 5784. ACIJournl. H. 1974. 28.E. 693700. “Shear strength of reinforced concrete beams. Nielsen. Vol. Draft version Dec. Taerwe. March 1957.Guide to EC2 Section 6 Leung. Polytechnic of Central London. J. PrEN 1990. R. 5. Deutscher Ausschuss für Stahlbeton. 535.C. (1976). Walraven. Berlin.
."Punching in reinforced concrete flat s1abs with holes". SINTEF Structures and Concrete. Pancaldi. Bulletin 146. 14. December 1979. Nylander. P. in CEB Bulletin 224. "Punching shear tests on reinforced high strength concrete slabs". 10. Petcu.4 Punching. R. R. 15. “Shear in Prestressed Concrete” pp. Structures Research Group.E.. Trondheim. Vo1. Deutscher Ausschuss fur Stahlbeton. Report 2.. B. American Concrete Institute.. J. July 1956. Proceedings of the international conference Concrete 2000. CEBBulletin d'lnformation 180. Andrade. 9. Birkhiiuser Verlag Basel. Detroit. Bericht Nr.. Marzouk. H. July 1995. 2835 5. Model Code Text Book. G.Y. KTH Stockholm.CEBBulletinNr. Stockholm 1960. 1980.. Edingburgh.B.S. Civil. A. Ostereichische Ingenieurzeitschrift. R. Ghali.Nr. ACIJournal.D. Elstner. EHT ZUrich. J. 17.E.. G. Polytecnic of Central London. J. Gomes. Base. M. "Safety and Performance Concepts". Pp. Hussein. 701713. "Punching strength predictions for twoway reinforced concrete slabs".C. 73053. pp. CEBBulletin d'Information 219. pp. Stanculescu. F..pp. Kordina. Transactions of the Royal Institute of Technology. w." TH Karlruhe. Regan. "Experimental investigation on the behavior of high strength concrete slabs" ACIStructural Journal."Mode1 Uncertainties concerning Design Equations for the Shear Capacity of Concrete Members without Shear Reinforcement". Civil Engineering 3. 21. Kinnunen. Vol. Pralong. Vol. . "Ultimate Limit State Principles". 2958. Tolf. Royal Institute of Technology. JanFeb. Marti.. September 1977. pp. J .. Xue.C.Comp Press. Birkhauser Verlag Basel. March/Apri11979. 1963 und 1964. 13.6882 8. Konig. Seible. Berich Nr. Tomaszewicz. Hognestad. pp. Ramdane. 185193."Der Durchstandswiderstand von Stahlbetonplatten nach neu ausgewerteten Versuchen". Moe.. Nr. "HighStrength . Thurlimann. M.. ACI Journal. Walraven. "Durchstanzversuche an Stahlbetonund Spannbetonplatten".. Regan P.R. Institutionen for Byggnadstatik.3 "Punching shear capacity of reinforced concrete slabs".. Hallgren."Shearing strength of reinforced concrete slabs and footings under concentrated loads" Development Department Bulletin d47. Report No. Kinnunen. "Shear capacity of reinforced concrete s1abs subjected to punching". 3. 1986 (in German) 20..14. G. P.57. "Shearing strength of reinforced concrete slabs".. "Behavior of high strength concrete slabs". STF70 A93082. 158. "Towards a consistent treatment of model uncertainties in reliability formats for concrete structures".SP2 Plates and Shells. 186192. Rumania. Institut fur Beton. Structural Mechanics. 1968. Stockholm (in Swedish with summary in English).. Regan. Technical Report 1994.PhDThesis. in fibPage 635 Table of contents .E.Guide to EC2 Section 6 Punching 1..6. 53. 49100."Punching resistance of reinforced concrete flat slabs". M.F.. No. 73052. Franz. Developments in Computor Aided Design and Modeling for Structural Engineering. 36 pp. 16. Nr. A. J. 6. of Structural Mechanics and Engineering. 18. Briindli. "Versuche an Stahlbetonkorper der Flachdecke im StiitzenbereichVersuchsreihen I en II. Gomes. 1960. Dept. S. Portland Cement Association. Polytechnic of Central London. P. Institut fur Baustik und Konstruktion. p 535. Kungilga Tekniska Hogskolan. K. A. Part 7. Tests on circular slabs". E. W.und Stahlbetonbau. "Punching of Concrete Slabs without Shear Reinforcement" Meddelande N. H. P. 88. ETH Zurich. 4.B. S.. D. No1ting. AlHussaini. 7. pp.. "Single legged stirrups as shear reinforcement in reinforced concrete flat slabs". Herzog. 1971. 1993. B. 1991 23... 19. Taerwe. 38. 79. Digler. H. pp.. Vol. "Effect of slab thickness on punching shear strength of concrete slabs. Sept. Nr. 14. 761773. A.A. “Punching of concrete slabs without shear reinforcement”. Fischer. Pralong. 6. U. Institut fur Baustatik und Kkonstruktion. Berlin.. No. V.216219. 371. Apri11961. No 2. Model Uncertainties. Revue Romaine des Sciences Techniques: serie de la mechanique applique. G. 11. H. Thurlimann.6871. 2.. 24. “Schubversuche an Stahlbetonplatten”.H.. 1.. Nylander. K. Stockholm. 12. L. 7. 1980 22. pp.. “Preassembled shear reinforcing units for flat plates”.
D. Nr. Year 7. 143149. 2000. Hanson. "Puncionamento assimetrico em lajes de concreto. J. Regan. pp. K... 37. 1970. 549568. Proceedings. 31. unpublished Report of Stahlton AG. 40. pp. March 1976. Birkhiiuser Verlag Basel. Stamencovic.E. 705716. Kordina. S. CEBBulletin d'Information No. Almeida. 5. Gerber..S.E.. Imperial College. 26. M."Shear strength of reinforced concrete flat PhD slabs without shear reinforcement".. Thesis Imperial College. Mau 1966. A. Regan. Vol. Journal of the Structural Division. ETH Zurich. pp.. December 1979 (in German). Shehata. Brandli.W. PCIJournal.. NovDec. N. prestressed with unbonded tendons".. N..L. 25. TU Braunschweig. pp. Jan. ASCE.. L.. 1974/75 (in German). Nov. J. J. 1972. v.J. COPPE/UFRJ. 1970.A. R... 1002 No ST3.P. P.Guide to EC2 Section 6 Bulletin 2.. N. Appleton. Burns.. Chapman.168.. W. 39.N.C. Report No.. Lucio. 27. 1970. 43. J. ACI Journal Vol.3. Yitzhaki. 899. London 1980. London. "Behaviour of reinforced concrete flat slabs". PhDThesis. Mast. N.L. "Ultimate limit state of punching in the (fib) FIP Recommendations for the design of posttensioned slabs and foundations". 1987 29."Tests on slabcolumn connections with shear and unbalanced flexure. A. July. "Shear reinforcement in reinforced concrete column heads". "Flat slabs with Stahlton column strip prestressing". 44.G. Nordisk Betong.E. Institute for Building Statics and Structural Design".. "Punching shear in reinforced concrete". Narasimhan. P. ACI Journal. pp.. "Genomstansning ay lift slabs (Punching of lift slabs)". J. Page 636 Table of contents . Report of the Institute for Materials. Oct.G. 1. Report Nr. ACI Structural Journal. Vol. Nov/Dec.. Thurimann. N. P . Pralong. Hawkins. 527542. "Stresses in flat plates near columns". Regan. London. V.. 63. J. September 1984 (in German). H. R. "Punching strength of reinforced concrete slabs"... Note to CEB Commission IV.M. Bao. 1971. 38. 4058. Yamazaki. "Shear and moment transfer between concrete slabs and columns". 36. "Ultimate strength tests of posttensioned flat plates". 1999. "Explanatory notes on proposed Model Code (MC'90)".73053.. 32.J.. 30. London 1971 34. May 8th 1990. 229252. 202223. Islam. Portland Cement Association."Tests on shear in slabs. Concrete Structures and Fire. No. Sept. P. Zurich.H. 33.. Mast. Structural Concrete. Anis.E. 86. L..768. pp. 2.902. Punching tests on reinforced and prestressed slabs". Nolting. A. 42.3."Plate stresses at columns near to free edges".E.89.W. Jan 196B. 1998. 41.. Park.1963. ACI Journal. Andersson. No. "Local strength of flat slabs at column heads". Journal of fib. Construction Industry Research and Information Association."Moment transfer from concrete slabs to columns". Construction Industry Research and Information Association. B. Rio de Janeiro. pp.E. Development Bulletin D129. Braestrup.1985. P . Structural Concrete. Hanson... 35. 39. Report Nr. 761.M. 28.. pp. Stahlton.A.6.
3 C7.9h z .eff Act h Ed ¨ hcr ¸ 3 3 ©2 ¹ After cracking this same bending moment.3 Crack control 7. the cracking moment can be determined by: N §h 1 · (7. must be taken by the cracked cross section: N §h 1 · §h · (7. SERVICEABILITY LIMIT STATES See example n.2) fct . SERVICEABILITY LIMIT STATES (SLS) 7.3.3.1 Minimum reinforcement concept For a rectangular cross section subject to combined bending and axial force.2 Minimum reinforcement areas SECTION 7.2 Formula for minimum reinforcement Formula 7.1 See example n.1 of prEN can be partially deduced by expressing the condition that the minimum reinforcement should be able to withstand the cracking moment working at a certain stress s fyk.Guide to EC2 Section 7 SECTION 7. 7. Figure 7.1 General considerations 7.3. together with the axial force.1 General 7.eff Act h Ed ¨ hcr ¸ As Vs z NEd ¨ 0.2 Stress limitation 7.1) M cr fct . 7.
9 D . ¸ 3 3 ©2 ¹ ©2 ¹ The lever arm z.8h for pure bending).eff Act 1 ª3 0. may be taken as a certain fraction D of h (around 0. Introducing the following notation: NEd z Vc Act bhcr Dh bh Equation above can be rewritten as: Vc ª 1 § h º· º fct .
8 for parameter D.4) fct . Factor alpha=z/h as a function of adimensional axial force Figure 7. With this assumption and for compression force or moderate tensile force. there is no compression block and under –0.03. Page 71 Table of contents . the values of kc may be estimated as: § V § h ·· k c 0.03 to –0. 1¼ » ¸ As Vs (7. since the deduction assumes that there is a part of the cross section which is in compression.7 ¸ ¸ (7.22 there is no need to provide minimum reinforcement according to the principle established above.2 Value of D as a function of the axial force It can be seen from the above figure that it is reasonable to assume a value of 0.22. Over bhfck 0.eff Act k c As Vs The above value of kc is valid only if D > 0.eff © © ¹ hcr can be computed as a function of the axial compression and the tensile strength of concrete.5) ¨ f hcr ¹ ¸ ct .4 ¨ 1 c ¨ 1 0.3) ¨1 ¨ f « ¬ hcr ¼ ¸ 3D © ct . In order to approximate kc in the following figure the value of NEd D=z/h is plotted against the adimensional axial force Q within a range of +0.eff ¬ ¹ (7.
eff Vc .Guide to EC2 Section 7 The cracking moment is given by the following condition: NEd M cr M bh 2 fct .eff 2 u 6 Vc cr2 u 6 M cr fct .
7) bh fct .6) bh bh bh 6 The depth of the tensile zone prior to cracking. relative to the centre of gravity of the cross section. (7. is given by the condition that the stress be nil when the existing axial force is applied together with the cracking moment: M Vc h 0 Vc cr3 12x x (7.eff Vc .
eff Vc . hcr can be calculated as: · § fct . 2 Therefore.
eff Vc . · Vc h h§ h 2¨ hcr x (7.8) ¨1 ¸ ¸ ¨ fct .
¸ hcr ¨ fct . Comparison of different approximate expressions for the calculation of the minimum reinforcement It can be seen that the 2 last methods give almost the same results.4. Value of kc as a function of the axial force In the next figure the minimum reinforcement for a section subject to a compressive force (expressed in terms of a reduced axial force) is shown calculated according to 4 different methods: .eff ¨ © ct . Figure 7.5). This procedure is on the safe side since it neglects the increase of the lever arm which occurs after cracking.eff of 2. .9 Mpa according to both the above formulation and the equation of prEN.Direct calculation (exact determination of D) .8. It can also be seen that the formula of prEN is mostly on the safe side.9) ¸ ¸» ¨ f ¸¸ « fct . Figure 7.8) in (7.Procedure explained above in which is taken as a constant equal to 0.3.Proposal of prEN .eff ¹ ¹ » © ¬ ¼ In the following figure the value of kc is plotted against c for a value of fct. the expression of kc becomes: ª § f Vc · · º V § k c 0.4 «1 c ¨ 1 1.eff (7. Page 72 Table of contents .4 ¨ ct .eff ¸ 2 2© ¹ © ¹ Introducing equation (7.Compensation of tension block.
EC2 [4] and prEN.1.Guide to EC2 Section 7 7. This assumption is on the safe side.3 Control of cracking without direct calculation C7.3. as will be seen. the crack width is given by: Vsr · I Vs § w k 2l t Hsm Hcm .1 Deduction of formula Using the MC90 formulation. C7. A more exact formulation could be obtained if eff were to remain a variable. This deduction. for which a certain stress s is achieved corresponds to the minimum steel ratio given by prEN equation 7.3.3. requires the introduction of simplifications and the assumption that the steel ratio.3.3. Formula for maximum bar diameter In this paragraph the formula for the maximum bar diameter is deduced for three crack width formulations: MC90 [3].
11) As Us .5k ' h d .6Ueff E s © Vs ¹ sr is the stress calculated for the fully cracked cross section for the cracking moment. (7. sr may be calculated by: f A k k fct .10) ¨ 1 kt ¸ 3.eff 2.eff ct c (7.eff kc k hcr Vsr ct .
Introducing the expression of sr into equation (7.k t ct.10).12) § · f k hcr 1.d . and rearranging: 3.5k' h .6 w k eff I= (7. k’ is a factor which is equal to 1 for bending and equal to 2 for tension.eff s 2. In the above expression.eff c ¨ ¸ s.
¹ © In order to obtain numerical values from this expression.0.12) can be written as: 720000 w k eff 1 I= (7. Therefore some assumption regarding eff must be made in order to obtain table 7.4.eff = 2.e. which is valid only for s > sr.38 hcr h 10  =5 (7.14) § 1 · s ¨ 1.1: k = 1. then formula (7.14) will give smaller bar diameters if the smallest the value of eff is used.88 ¸ s. is a function of 3 parameters.5k ' h d .3 mm With the above values equation (7.eff c (7. The assumption which will be made is that eff = eff.eff s ¹ © The above expression. This assumption is justified because lower values of s will not produce cracking and because if cracking does occur.13) h.d 2 h fct.9 N/mm2 wk = 0. s = sr).min ct .2. The value of eff can therefore be taken from the following equation: f kk hcr Ueff Ueff .15) Vs 2. k’ = 1 (pure bending) kt = 0.1.3 (assumption on safe side) kc = 0. it is necessary to assume values for those coefficients which are not considered as variables in table 7.min (i. The following values have been assumed to derive table 7.0 h 0.
16) I  V2 2.12). this equation can be simplified into: 3.5k ' h d .6 w k fct .eff k c k Es hcr 808258 (7. If the above value of eff is substituted into Eq. (7.
V2 1 kt .
1.Minimum reinforcement ratio for a given value of s > sr.4. MC90 .14) is given for each stress level. s s In Table 1 the minimum reinforcement ratio which can to be used in equation (7. Value of corresponding Imax Page 73 s which fulfils the condition Table of contents .3 mm and that included in MC90 Table 7. the maximum bar diameter obtained using equation (7.14) for wk=0. Table 7. In the same table.3 are also given.
5. the crack width is given by: § § sr · · I · s§ ¨ w k = 1. Theoretical and actual values of Dmax. according to MC90 Using the EC2 formulation. Figure 7.Guide to EC2 Section 7 These results are plotted in the following figure.7srm H sm .Hcm .
eff A ct k c k fct.25k1k 2 (7.1 2 ¨ ¸ ¸ eff ¹ E s © © © s¹ ¸ ¹ sr is the stress calculated for the fully cracked cross section for the cracking moment.18) sr = As 2.d . sr may be calculated by: fct. = 1.7 ¨ 50 + 0.5k' h .17) ¸ ¨ 1.eff k c k hcr = (7.
1 2 ¨ ° © s.eff Introducing the expression of sr into equation (7. s.eff s 2.7 s § fct.50 ¾ 2 § · 1.10).eff k c · ° ° 0.d .19) I=® .25k1k 2 hcr ¸ ¸ ° ¨ 1.5k' h . and rearranging: ½ ° ° ° ° Es wk ° ° eff (7.
d 2 h fct. ¹ ¸ °¨ ° © ¹ ¯ ¿ In order to obtain numerical values from this expression.50 0.22) ° 1.50 ¾ 10 eff I=® 2 § § 1.eff s ¹ ¸ ° ¹ ¯© ¿ Similarly.2: k = 1.¨ ° ° ¨ © s. it is on the safe side to assume for this purpose that s = sr and that therefore the value of eff can be determined using equation (7.10 With the above values equation (7.4.50 ¾ (7.eff = 2.5 N/mm2 E1E2 = 0. equation (7. The following values have been assumed to derive table 7.0 h 0.25 k1k2 = 0.1 2 .15).19) can be rewritten as: wk ½ f kk 70588 ½ 20 Es hcr ° ° I=®  ® . As stated before. With this assumption. a value must be given to eff in order to obtain values from this expression.50 ¾ ct.2.414 · · s ° ° ¸ ¸ ° ¨ 1. it is necessary to assume values for those coefficients which are not considered as variables in table 7.19) can be written as: ½ ° ° ° ° wk 117647 ° ° (7.eff c .21) . k’ = 1 (pure bending) hcr h 10  =5 (7.20) h.3 (assumption on safe side) kc = 0.
1.7 s ° 0.5k' h .d .25k1k 2 2.
Page 74 Table of contents max . ¯ s ¿ s ¯ ¿ 2 In the following table. the results obtained by application of the above formula and the values of according to EC2 are compared.
according to EC2 Using the prEN formulation. the crack width is given by: § § sr · · I · s§ w k = 1. Value of corresponding Imax s which fulfils the condition In the following graph.Hcm . Fig 7.2 EC2 .7srm H sm .6. the above results are plotted.Guide to EC2 Section 7 Table 7.Minimum reinforcement ratio for a given value of s > sr. Theoretical and actual values of Dmax.
5k' h .425k1k 2 (7.k t ¨ ¸¸ ¨ ¸ eff ¹ E s © © © s ¹¹ sr is the stress calculated for the fully cracked cross section for the cracking moment.23) ¸ ¨ 1.24) sr = As 2.eff k c k hcr = (7.d .eff A ct k c k fct. sr may be calculated by: fct. = ¨ 3.4c + 0.
d .eff s 2.5k' h .3.eff Introducing the expression of sr into equation (7.23).25) I=® . and rearranging: ½ ° ° ° ° Es wk ° ° eff (7.eff k c · ° ° 0. s.4c ¾ 2 § · s § fct.425k1k 2 hcr ¸ ¸ ° ¨ 1.k t ¨ ° © s.
4. k’ = 1 (pure bending) hcr h 10  =5 (7. ¹ ¸ °¨ ° © ¹ ¯ ¿ In order to obtain numerical values from this expression.40 0.2.9 N/mm2 kt = 0.0 h 0.eff = 2. it is necessary to assume values for those coefficients which are not considered as variables in table 7.26) h.25) can be written as: Page 75 Table of contents .d 2 h fct.425 k1k2 = 0.3 (assumption on safe side) kc = 0.17 C = 25 mm With the above values equation (7. The following values have been assumed to derive table 7.2: k = 1.
928 · · s ° ¸¸ ° ¨ 1.4c ¾ ct.Guide to EC2 Section 7 ½ ° ° ° 200000 ° wk I=® .28) ° 1.eff s ¹ ¹ ¿ The above expression is a function of 3 parameters and again can be simplified with the assumptions explained above.3.27) ° § § 0.¨ ° ¨ ¸ ¯ © © s.88 eff (7.85 ¾ 5.85 ¾ I=® k  ® (7. into: w Es ½ f kk 100000 ½ 13.k t .eff c .65 hcr ° ° .
s ° s 0.5k' h .d .425k1k 2 2.
7. Comparison of the 3 models considered Page 76 Table of contents . Table 7. Theoretical and actual values of Dmax.3. according to prEN The three formulas analyzed are compared in the following graph. It can be seen that good agreement is obtained between theory and prEN table. The values included in the tables of the codes do not always match the theoretical equation but are generally conservative for EC2 and prEN.3 prENV .3. It can be seen that the theoretical results for EC2 and prEN are very similar while the MC90 equation is somewhat more conservative. Figure 7.Minimum reinforcement ratio for a given value of s > sr.8 Theoretical and actual values of Dmax. s ¯ ¿ s ¯ ¿ This curve is represented in the following figure and given in numerical form in table 7. Value of corresponding Imax s which fulfils the condition Figure 7. The small differences observed are due to the need to use commercial bar diameters in table 7. The MC90 table values are somewhat above the theoretical curve as already shown before.
9 0.5 in case of EC2).5h h . then the value obtained from the tables must be corrected by the following factor: fct.29) 2.d .4 0.1h and also that a problem of pure bending is being analysed and therefore.9 N/mm2 (2. Furthermore.1h k' fct. kc = 0. If different values are assumed for these parameters.Guide to EC2 Section 7 7.3.5h.3. hcr = 0.eff k c hcr 0.eff k c hcr 1 Part of cross section compressed: = (7. the tensile strength of concrete has been assumed to be 2.4 and k’=1.2 Correction for cover In the above expressions it has been assumed that h – d § 0.
1 2.9 2 h .d .
eff k c hcr 1 = 2.9 0.4 0.1h 1 fct.5h h .eff k c hcr 0.d . 1 Cross section in tension: fct.
d .9 2 h . k' 2.
eff c cr part of section compressed 2. 2 (7.d .9 2 h .30) It can therefore be written that: f kh Is = Is* ct.
eff k c hcr all section in tension 2.9 4 h .31a) (7.4 Calculation of crack widths (7. Is = Is* 7.3.d .31b) fct.
3. Beeby [1.3. it has been tested against an experimental data base including results from the researchers Rehm & Rüsch [1113]. shows that for all 3 models. Falkner [6] . The criteria for the selection of the experimental results are clearly explained below. This fact suggests that in future editions of EC2. under the relevant combination of loads.2] and Jaccoud [8].1 Introduction The formula proposed for crack width is a mixture of EC2 and MC90.4.Hcm) may be calculated from the expression: fctm 1+ D eff . C7. etc. It is the intention of the authors to avoid ambiguity and provide a self explaining instrument which can be used by other researchers in the future so that the work carried out here need not be repeated. Formula for crack width 7. 7. Krips [9] . Hartl [7] . the error margin grows as the crack width grows and that all models tend to underestimate the crack width when it is large. however. the design crack width can be determined using the following expression: wk = srm (Hsm – Hcm) where wk design crack width srmax maximum crack spacing mean strain in the reinforcement.2 Proposed formulation According to the prEN proposal. In order to evaluate the accuracy of this formula.4. The performance of the new prEN formula is also compared to that of MC90 and EC2. Elighausen [5] . After a review of the proposed model. the results of the comparison between model and experimental data are presented. The results show small differences between the 3 models.3. some correction might be needed to allow for this situation. A detailed presentation of this data base is therefore given. taking into account sm the effects of tension stiffening. The data base has been drafted specifically for this document and is detailed in appendix A.4. The analysis. mean strain in concrete between cracks cm The strain difference (Hsm . This fact is unfortunate since the control of cracking in normal structures is most important when cracks are large.
kt s eff (7.Hcm .6 s Hsm .32) t 0. s .
= . s fctm 1.k t H sr = Es Es Es where stress in the tension reinforcement assuming a cracked section.D eff .
eff is the area of concrete surrounding the tension reinforcement of depth.5(hd). where hc. this simplification makes it easier to apply the model in practical cases and does not imply any significant loss of accuracy as is shown below. sr eff  Hsr H sr = This is a simplification which is exact for pure tension but not for Es Es bending. hc.eff effective tension area. Page 77 Table of contents .ef is the lesser of 2. De ratio Es/Ec A Up. Ac. However. (hx)/3 or h/2 (see figure).ef .eff = s Ac Ac.
effective tension area for upper surface C .4c + 0. k2 = 0.effective tension area B .8 for high bond bars k1= 1. k1 = 0.4 for long term loading b) Slab a) Beam A .Guide to EC2 Section 7 kt factor dependent on the duration of the load kt = 0.33) eff where I bar diameter.425k1k 2 (7.0 for pure tension For cases of eccentric tension or for local areas.6 for short term loading kt = 0. c cover to the reinforcement k1 coefficient which takes account of the bond properties of the bonded reinforcement.6 for bars with an effectively plain surface k2 coefficient which takes account of the distribution of strain.effective tension area for lower surface Figure 7.9 Definition of the effective tension area In situations.level of steel centroid B . where bonded reinforcement is fixed at reasonably close spacing within the tension zone (spacing 5(c+ /2). the maximum final crack spacing can be calculated from the expression: I srmax = 3.effective tension area c) Member in tension B . intermediate values of k2 should be used which can be calculated from the relation: H + H .5 for bending k2= 1.
the cover.34) 2H1 Where 1 is the greater and 2 is the lesser tensile strain at the boundaries of the section considered. with regard to crack spacing. In EC2. Page 78 Table of contents . c is implicitly taken as 25 mm. c. This suggestion is backed up by the following figure. is introduced explicitly into the expression of the crack width as suggested by A. Beeby. assessed on the basis of a cracked section. As can be seen. taken from reference [1]. k2 = 1 2 (7. which clearly shows the dependence of the crack spacing on this parameter.
7 2 25 + 0. first Im will be applied to the equilibrium equation (7. In the case in which 2 different bar diameters are used.33) to derive Ieq and then Ieq will be derived by considering 2 diameters in the determination of the steel area (7.36) of prEN.7 times srm.4c + 0.10 Influence of cover on the transfer length acccording to Beeby [1] Also.37) (7.38) fctm A ctm = WS nI1 + nI2 .35) of EC2 becomes 3.34).Guide to EC2 Section 7 Fig. In EC2 [4] it is recommended to use the average diameter Im. on a formal level. an equivalent diameter Ieq has to be determined in order to apply the above formulation. MC90[3] and prEN [10] suggest for this case the equivalent diameter Ieq. the formula of prEN gives srmax instead of srm.7.25k1k 2 p.35). I 1. srmax is obtained as 1. fctm A ctm = W S n I lt (7.33) changes in case of the use of 2 different diameters into (7.eff (7.425k1k 2 I p. although MC90 provides no definition of eq. (7. The definition of Ieq depends on the definition of eff.eff (7. Thus. To show the difference.
lt Hence: lt = fctm A ctm WS nI1 + nI2 .
6 and.80 WS n1I1 + n2 I2 .39) since srmax = 2lt and srmax = I/3. (7. it can be written: srmax = =2 SI W 1.eff fctm 1 = .eff Ieq 2fctm A ct. according to MC90[5] A ct.
36) can be seen that the equivalent diameter is: n I + n I .80 × 4 eq 4 From (7. 1.
Ieq = 1 1 2 2 = Im n1 + n2 This is correct when the steel section in = If this is not the SI2 As is defined with A s = n1 + n2 .
S 2 n1I12 + n2I22 . eq A ct.eff 4 and the steel area is instead (7.41) case.40) (7.
A c. SI1 SI2 2 4 A s = n1 + n2 then.eff = and therefore 4 4 srmax 2 2 fctm n1I1 + n2 I2 .
=2 W n1I1 + n2 I2 .
38) can be derived that the equivalent diameter is: Ieq = n I n I 2 1 1 1 1 2 + n2 I2 . described by (7.42) From (1.
+ n2 I2 .
(7.eff. For this reason the definition of Ieq is made explicit in prEN. since the code user will naturally define eff as As/Ac. Page 79 Table of contents .43) It can be seen that EC2 and MC90 could be misleading and thus should be clarified.
or concrete qualities less than 20 N/mm2 and steel qualities of less than 400 N/mm2. For this purpose. P.4 Analysis of the experimental data This paragraph includes the analysis of the experimental data as well as the comparison between the values obtained from the experimental data base and the values obtained from the theoretical models.For the determining of the crack spacing the number of cracks present at the last phase of the test is always considered since it is the closest to stabilized cracking. since theses test can be representative of walls subject to shrinkage and temperature.32). since this analysis provides some interesting conclusions.3. However.Guide to EC2 Section 7 7. only results within a stress range in steel from 150 to 350 N/mm2 were considered for tests involving direct actions. This expression is also plotted in Figure 7.4. . also for the maximum crack width.The materials used in the tests need to be similar to the materials used today in the building of structures. The distribution function of the maximum crack widths shows clearly a tendency to a normal distribution.1 Probability function of the maximum crack width In this paragraph a first analysis of the probability distribution of the maximum crack width is made.4. in order to verify a satisfactory performance of the new proposal. Experimental data generally includes the values for the mean crack width and.The stress range should be a serviceability range. the data base includes data from different researchers. The comparison includes not only the prEN model. Equation (7. . first an analysis of the distribution function of the maximum crack width is made. ¨ 1 f x. This rule leads to discard tests which use low bond rebars. The data has been selected among the different tests available following a few simple rules: . Figure 11 shows the crack width distribution function for all selected members within the service stress range. which is the crack spacing given in equation (7. steel stresses up to the yielding stress of steel were considered. in most cases. For indirect actions. The models are compared on the basis of the mean crack width because it is difficult to determine the experimental characteristic crack width. For this the results of Rehm & Rüsch [1113] are used. since in many of the experimental results the maximum crack width is also available.3 Presentation of data base As stated above. .4.44) shows the mathematical equation for the normal distribution function.3.3. but also those of EC2 and MC90.4. 7.11 showing very close agreement with the experimental data. 7.
P . = e© 2S § x.
Figure 7.7.2 · ¨ ¸ 2 2 ¸ ¹ (7. 250 N/mm2.11. 300 N/mm2 and 350 N/mm2. The division of the results into different stress levels shows the evolution of the maximum crack width as a function of the stress level. It can be stated that for normal durability conditions.12 (distribution function) and Figure 7.44) Fig. It can be seen that for a stress level of 200 N/mm2 there is a probability of 95% that a maximum crack width smaller than 0. This is consistent with the available experience. Maximum experimental crack width distribution Having proven that it is reasonable to assume that the probability distribution of the maximum crack width can be assumed is a normal distribution. stress levels under 250 N/mm2 will not pose any Page 710 Table of contents . These functions are displayed in Figure 7. the distribution and the density functions were evaluated for each stress level within the service spectrum: 200 N/mm2.13 (density function).3 mm occurs.12 shows the distribution of the crack width for the different stress levels.
10] is presented. which crack opening is typical for a certain stress level. Fig.25 mm for 250 N/mm2 and so on. 7. It can be seen in Figure 7. that in the given data. Density curves of the maximum crack width for different stress levels This shows. all the data obtained by the evaluation of the experimental results and all the data obtained by processing the corresponding input results according to the standards. MC90 and PrEN Page 711 Table of contents . since these are the values they assume. is displayed. This comparison shows the performance of the formulae.4. not only against each other. Since prEN and MC90 provide the characteristic crack width. but also against the test results.13. to EC2. Figure 14 Comparison testcalc. acc.4.7 and 1.3. 0.Guide to EC2 Section 7 serious cracking problems Figure 7.5 respectively. Normal distribution of the maximum crack width for different stress levels The density curves show the concentration of the crack opening..19 mm is typical for a steel stress of 200 N/mm2.7.14. a maximum crack width of 0. the mean crack width has been estimated by dividing the value given by these codes by 1.12.13.5 Comparison of the standards In this paragraph a direct comparison between the formulae according to [3. In Figure 7.
005 mm (overestimation).13 mm. A better view of this information can be obtained by plotting the error of the estimation instead of the crack value. It can be seen that good agreement is found and that the error can effectively be assumed have a normal distribution.4. With a 95% probability the underestimation will be less than 0. but the values of the 3 models are fairly close). Page 712 Table of contents .16 Distribution function of the error in the estimation of the mean crack width The diagram below shows the density function of the errors of each model. its standard deviation and the 95% confidence intervals (assuming a normal distribution for the error) For example the mean error according to EC2 is werror = 0.Guide to EC2 Section 7 All 3 formulae perform quite closely to each other.exp (7. Errors in crack width prediction The above table shows the mean value of the error. The error is defined as: wm – wm. A normal distribution was assumed based on the following figure which shows the experimental error compared to normal distribution having approximately the same mean value and standard deviation (the theoretical value of the curve corresponds to EC2.45) This result is given in the following graph and table: Figure 7. Figure 7.12 mm and the overestimation less than of 0. This is shown by the corresponding trend lines.15 Error in crack width prediction Table 7.
8). EC2 [4] and MC90 [3] and the PrEN [10]. the crack opening is expected to be critically larger and the prediction more important than for small cracks under lower steel stress.4 c 0. the crack formulae [3.17 O » Es ¬ Vs ¼ ¬ Us ¼ The EC2 cracking opening formula is: w k which can be used as a design formula. and the selection criteria given above.Nevertheless it could be verified that the existing models.46) «1 » « k 3 c k 1k 2 k 4 O » Es ¬ Vs ¼ ¬ Us ¼ Assuming the prescribed values k3=3. for given b. three things became manifest: .425 and considering the bending case (k2=0. The nondimensional depth of the neutral axis is 1 [2 D e Us 1 E .cr º ª I º (7. d.Guide to EC2 Section 7 Figure 7.10] increases for larger crack openings. This was confirmed by the statistical evaluation. . we want to deduce the steel reinforcement amount As and its design tension Vs in order to have a crack width wk lower than the fixed value w k . because especially for higher steel stresses. which demonstrated that below 250 N/mm2 the anticipated maximum crack opening will not exceed a value of 0. b.076 mm). h.cr º ª I º wk (7. that the PrEN has a good performance range and will be an adequate substitute for the existing EC2 formula.063 to 0. Also all models tend to underestimate the crack width when it is large. The mean value of the observed error is in all cases close to zero.4.3.7.1 Exact derivation Vs ª Vs. d’. This is a critical point.3. With all this information it can be said.4 mm with a plausibility of 95%. Also the standard deviation of this error is relatively small (0. provide acceptable predictions.4. In particular.4. During the process of gathering and computing all the result data. k4=0.4. It might be interesting for future proposals of the EC2 crack prediction formula to provide an adjustment to compensate this tendency.5) with high bond reinforcement (k1=0. and fixed M.10] proposed for the crack prediction for the 3 different models discussed above was evaluated. 8 7.Within a steel stress limit of 250MPa the crack width does not have to be necessarily checked.47) «1 » «3.4. it results Vs ª Vs.The error of the formulae in the evaluation of the crack width [3.17 Density function of the error in the estimation of the mean crack width 7.6 Conclusion With the experimental data gathered. .
[ D e Us G E G ' .
0 (7.48) 2 The steel tension is D e Q G [ .
fctm k t Vs 2 2 (7.49) 2 ª3n Us ª G [ .
E G ' [ .
50) [2 2D e ª 1 E . º [3 º ¬ ¼ ¬ ¼ Q M M0 cr M b h 2 we get k t fctm 6 Assuming Us (7.
[ G EG ' º ¬ ¼ Table of contents (7.51) Page 713 .
Guide to EC2 Section 7 G [ .
E G ' [ .
2 D e Q G [ .
2[ 3 p 2 3[2 p G EG ' 1 E .
17 Us Es w k setting w 0k k f t ctm p (7. [ where p=Vs/(ktfctm) (7.4 c 0.52) When w w k .53) Combining the above equations it results ª º DeQ G[. after some calculations we deduce w0 k O n Us IO 3.
2D O w0 [2 k e2 ªGEG ' 1 E.
[º De » u « ¬ ¼ 2 2 2 [ ª º ¬ 3.34De IO ¬GEG ' 1 E.4 c [ 0.
[¼ ª º » ¼ ¬ G[.
E G '[.
¼ « ª º 3[2 2[3 » u« « ªGEG ' 1E.
[º ª G[.
2 E G '[.
20 e Q º p Q * >17D e u 1 5u 2 @ 0 (7.2 Approximated derivation The procedure discussed above is quite laborious as it requires iteration.2 º » ¬ ¼ ¬ ¼¼ ¬ (7.56) The general formula for w wk Vs Es w k gives ª O § D e Us · º § IO · ¸ ¨1 ¸ » ¨ 3.9d constant and independent from [.3. it is possible to set the tensional level Vs.9.60) Q * p p2 De Q that defines the maximal bar diameter. solving for I we obtain ª17c( Qp D e Q * ) 5Q * w ok º ¼ Imax ¬ (7.55) « O ¬ (p D e )Us O ¼ 7. gives the neutral axis position from which the reinforcement tension and its amount can be determined. VsAs0.5 practically impossible for bending problems.5·(1G) and then reevaluating [. easier to be applied.185Q Q the following equation is obtained D p 2 5 Q * ª3. Us from (7. as the parameter p is defined. Alternatively. An alternative procedure. which satisfies the cracking limit state corresponding to a fixed value of the tension in the steel Page 714 Table of contents .57) «1 Us p © O ¹¼ © Us ¹ ¬ For further simplification of the problem. for a given value of Vs. the neutral axis is calculated from (7.4c 0. If it is not the case.52). In this case.Therefore. it is necessary to set in the = 2. which gives º Us ª 5.185Q/(pG) (7. The procedure. for example equal to the limit one.17 (7. assuming G=0.9d=M and Us=0.4. therefore = 0. being the value =0.243 and by definition Q Q c w 0k Q* u1 u2 1.7. In this case too.18 GO (7.58) 1 1 I I 0. and to evaluate the corresponding reinforcement amount Us and the maximal bar diameter. is based on the assumption of a lever arm h0 = 0.59) « » ¬ ¼ This formula is easy to solve and leads to the desired values Us and Vs.4u1 Q Q * 0. aimed to the determination of the reinforcement amount and its tension corresponding to fixed crack width values and stress level.50) and the maximal diameter can be derived solving with respect to I.54) which numerically solved.88Us w ok Imax 2c » (7. requires to assume the value of the bars diameter I.
The law to calculate the reduction of stiffness is deduced from equation 7.1 General considerations 7.1 Instantaneous deflections Instantaneous deflections are computed by applying the total load on a structure in which there is a reduction of the stiffness.4.Guide to EC2 Section 7 7.] .4 Discussion of the general method followed for deflection calculation 7.3 Checking deflections by calculation C7.4 Deflection control 7.4.4.4.8 of prEN 19921: 1 M M M § 1· § 1· = = ] ¨ ¸ + 1.2 Cases where calculations may be omitted 7.
] . ¨ ¸ = ] 1.
(7. the following relationship is obtained: II III Ie = (7.61) r EIe EII EII © r ¹II © r ¹I From this equation.] .62) ]II + 1.
In both cases.4.]2) .3.]3) + fg2+q(t2.4.2 provide slenderness limits for lightly reinforced and heavily reinforced concrete elements.4.2 Deflections due to creep Complex Loading History The interpretation of prEN 19921 regarding deflections due to creep is not clear when the loadhistory is complex.64) 1+ M The question then arises about which time should be used to compute the creep coefficient. A typical load history for buildings could be: • Application of self weight at 10 days • Application of the remaining dead load at 60 days • Application of quasipermanent load at 365 days.fq1(t2.4. • Compute fg1+g2(t3. that part of the deflection due to g1 occurs at time t2 due to the reduction of stiffness produced by the application of load g2.]3) + fg1+g2(t2. • Compute fq1(t2. The creep of this extra deflection must therefore be referred to time t2.eff.]3) .f = fg1. This expression takes into account.t2) and ]1. ]1) with (t.]) . Also.]1) + fg1+g2(t2.14 and MC90 in table 7. for example.2.4.f = fg1. ]2) with (t. To solve this problem the following procedure is proposed: • Compute fg1.1 Assumed Load History The load history influences the value of the deflections. In this study a realistic load history has been taken into account.63) II 7.]1) + fg1+g2+q(t3. As described in paragraph 7.4. • Compute fg1+g2+q(t3. • Compute fg1+g2(t2. 7.t2) and ]2. for a simply supported beam the Page 715 Table of contents .fq1(t2.]) + fg1+g2+q(t3. II and III are calculated using Ec.3 Parametric study of slenderness limit 7.fg1+g2(t3.]3) + fq(t3. ]3) with (t.1 Introduction ENV 199211:1991 in table 4.t3) and ]2 • Compute the total deformation by using the following expression: fg1 +g2+q.2.e. the assumed load history involves 3 dates for the application of the loads: • Application of self weight of the structure. g1 at time t1 • Application of remaining dead load g2 at time t2 • Application of quasipermanent live load \02q at t3 prEN 19921 proposes to take into account the creep of concrete by using an effective modulus for concrete: 1 Ec.(t1.]3) (7.t3) and ]3.5. ]2) (7.1.ts).]3) .65) This complicated and timeconsuming procedure is necessary due to progressive cracking of cross sections.2. ]1) with (t. since no reduction of stiffness occurs after the application of g1+construction load: fg1 +g2+q. ]2) with (t.66) 7.t1) and ]1.2. This is what is achieved by the above expression.(t1. In case a construction live load equivalent to the value of g1+g2+ 01q is assumed to be applied at time t1. M(t.2 Longterm deflections 7.3 Deflections due to shrinkage Deflections due to shrinkage are computed for a stiffness corresponding to the quasipermanent load condition. III The coefficient applied to the stiffness of each section is obtained as: I coef = e (7.(t1.fg1+g2(t3.eff = (7.4. the above expression is greatly simplified.(t1. taking into account an effective modulus with a creep coefficient corresponding to the start of development of shrinkage (i. ]3) = fg1. 7. end of curing).
60 Q tot  G2 = 0. is defined as the ratio of tensile reinforcement As. required reinforcement. • Distribution of reinforcement. .45 Q tot G2 = 0. live load: g1 = 45%qtot. Slenderness limits are calculated by limiting the total deflection to L/250. considers the influence on the slenderness limit of the following parameters: • Complex load history. These values are given for a steel yield stress of 400 N/mm2.075 Q tot For flat slabs. typical values for these relations could be: . The present proposal for section 7.4 of prEN 19921. The influence of this parameter on the slenderness limit has been studied for relative humidity varying from 50 to 80%. According to Spanish practice. It has been investigated whether the limitation of the deflection producing cracking of partitions to L/500 can be more restrictive. Reinforcement in real beams is not constant. The parametric study. The cover has been assumed as 1/10 of the total depth. respectively.3. There have been complaints in the sense that this table is too conservative.30 0.Distribution of reinforcement is considered constant over the beam length. The influence of the real distribution of reinforcement on the slenderness limit has been studied. The slenderness limit curves which are presented in the following paragraphs. Page 716 Table of contents .S.Concrete Strength: 30 N/mm2 . 7. considers a large range of variables affecting the deformation of concrete structures.4.25 = 0.30 0.Tensile and compressive reinforcements are those strictly needed for ULS. . or too general. For each part of the study.Load history: t1/t2/t3 =10/60/365 days .20 = 0. to effective cross section bd. is based on this study. are given for different reinforcement ratios. The reinforcement ratio. The parametric study described in the following sections. The reference values for which the study is formulated are the following: . the slenderness limit is defined as the relationship between the span and the effective depth L/d. L/d=18 and L/d=25.06 Q tot For this study. • Concrete grade • Percentage of self weight (g1). g2 = 30%qtot and q = 25% of qtot. This means.Permanent load vs. t2 and t3 on the slenderness limit has been studied. analysis has been studied in order to take into account the roundoff in detailing. control of deflection which produced cracking of partitions (referred to in this document as active deflection). The influence of the values of t1. . in order to quantify their influence and study the possibility of including them in the calculation of the slenderness limit. G1 = 0. • Influence of relative humidity. that the equivalent values for a 500 N/mm2 steel would be L/d=14 and L/d=20.20 Q tot \ 02Q = 0. • Real reinforcement vs.30 Q tot \ 02Q = 0. G1 = 0.2 Assumptions for parametric study The parametric study which follows has been carried out for a simply supported beam with a cross section of 100 x 30 cm2. and are inversely proportional to the steel grade. additional deal load (flooring and partitions g2) and quasipermanent live load ( 02q) with respect to the total load (qtot). as for prEN 19921. The relative humidity affects long term deflections through creep and shrinkage.Relative humidity of 70% .Steel Yield Stress: 500 N/mm2 . one of the above parameters is varied while the others remain constant. The effect of considering a 5% to 10% increase in the real reinforcement with respect to the required reinforcement determined from U. • Control of total deflections vs. which has been carried out according to the procedure described above.Guide to EC2 Section 7 corresponding values are.Quasi permanent live load is 30% of characteristic live load .For oneway slabs.L.
5 0. qULS. The reference load history is: t1/t2/t3=10/60/365 days Two other load histories are considered: t1/t2/t3=7/14/365 days and t1/t2/t3=28/90/365 days.Calculation of the ultimate bending moment (MULS).4. compressive reinforcement is provided and taken into account in the calculation of deflections.72 qULS . Influence of load history on the slenderness limit.45 + 0.25 qtot = 1. the influence of the load history is very limited. This suggests that simplifications are possible.5. . 7.18.4. Slenderness ratio for two rectangular cross section of different dimensions 7.The values of g1.1. Figure 7. As can be seen no significant difference can be observed. the procedure is repeated until convergence is achieved.5%.38 qtot qtot = 0. When compression reinforcement is needed in order to yield the tensile reinforcement.2. This cross section is rectangular of dimensions b x h = 30 x 50 cm2. If the deflection obtained is not L/250 for total deflection or L/500 for active deflection.35 (g1 + g2) + 1. all other parameters being those taken as reference (see section 3.5 Influence of load history The load history used for the parametric study is described above in 2.The ultimate load.5. As can be seen.2).30) qtot + 1.A certain span length.3.Guide to EC2 Section 7 7. This provides an upper bound estimation for total deflection (not so for active deflection). g2/qtot and q/qtot. In order to insure that the particular dimensions of the cross section are not important. assuming a simply supported beam: M qULS = ULS 8 L2 . Figure shows the comparison between both rectangular cross sections. It is assumed that the construction live load is applied at the same time as the self weight so that the section is fully cracked from the beginning. a rectangular cross section of b x h = 100 x 30 cm2.3. is assumed. The calculation of the slenderness limit has been carried out in this case for two steel ratios: 0. L.t2.5% and 1.(L/d)(t1. ((L/d)(10.5 q = 1.3.1. is determined from the MULS.365). the slenderness limit for different reinforcement ratios has also been determined for the cross section used by Beeby in [2].The total service load (qtot) is determined from qULS. .t3))/ (L/d)(10.The deflection is computed.60.3 Method for determining the slenderness ratio In order to determine the slenderness ratio. According to the general method of prEN 19921. .60. g2 and q are determined from the above ratios. as stated above.35 (0. Table 7.4 Influence of the dimensions of the cross section used The cross section assumed for the parametric study is. according to: qULS = 1. the following steps were taken for each reinforcement ratio: . and the assumed ratios for g1/qtot. One such simplification. consisting in considering a single time of loading together with an equivalent creep coefficient is described in detail in section 7. The results are shown in Table 7.4.4.4.365) Page 717 Table of contents .
Outside this length minimum reinforcement was considered.20. Figure 7.5%) and assuming a length (ls) of the supplementary reinforcement of 60.3. the limit of active deflection to a maximum value of L/500 is a more strict condition.6.7 Influence of distribution of reinforcement Distribution of reinforcement in beams is not constant in practice. The results are given in Table 7.3. Normally. Influence of the length of supplementary reinforcement on the deflection of a simply supported beam.19. The influence of the length of the additional reinforcement has been studied for two reinforcement ratios (0. Figure 7. Influence of round–off extra reinforcement in the deflection of a simply supported beam 7. 7.8 Active deflection vs.4.4 of prEN 19921.Guide to EC2 Section 7 7.6 Influence of additional reinforcement The slenderness ratio is referred to the deflection occurring in case the strict reinforcement is placed in the beam. total deflection Table 4.5% and 1. 7. However.20 – Slenderness Limit for Total and Active Deflection Page 718 Table of contents . The influence of this factor has been studied by comparing the deflection of a beam with different required reinforcement ratios in case an extra 5% or 10% reinforcement is provided in tension. 7.14 of ENV199211:1991 as well as table 7. as can be seen from figure Fig. Table 7. Further consideration should be given to this fact. on the safe side. of the required values.6 by comparing the deflection for ls=100% to the deflection for the different values of ls. it is normal for structures to have extra reinforcement due to rounding off. 90 and 100% of the span. The results are given in Fig. It can be seen that for low values of the reinforcement ratio and no influence is detected. 80.3. minimum reinforcement is placed near the points of zero bending moment (base reinforcement) and a supplementary reinforcement is placed in the areas near the maximum bending moment. However. has been determined by limiting total deflections. The difference becomes significant only for high reinforcement ratios and small length of additional reinforcement.19.4.4.
different load distributions have been assumed. are obtained approximately for a ratio of qcp/qULS=0.21. in the end.7.7. then by fixing different values for the quasi permanent service load (qcp)/ultimate load (qULS) ratio.10 Influence of the percentage of self weight. It is also interesting to note that the values of l/d = 20 for 0.3. Page 719 Table of contents .3.22.5 g2. If it is assumed that g1 § 1. significantly different values for the permanent to live load ratios for the quasi permanent combination. 60. superimposed dead load and live load with respect to total load. The slenderness limit was determined for these two steel ratios and relative humidity of 50.5 q As explained in section 7. can be taken as 0.5 times the ultimate limit state (ULS) load. the different ratios of qcp/qULS considered and the resulting ratios of g1/qtot. g2 and q may be determined for each ratio. 7.4. 7. qcp. For this study. superimposed dead load and live load to total service load ratios The comparison of the results of the slenderness limits for the different quasi permanent service load to ultimate load ratios is plotted in Fig. Influence of relative humidity on Slenderness Limits for Total and Active Deflections (referred to 70% RH) From Fig. since only 30% of the live load is considered quasi permanent. 7.5% and = 1.4. the ultimate load qULS is known from the value of the reinforcement. This provides. g1. Total service load/Ultimate load ratios and corresponding values of self weight.9 Influence of relative humidity The influence of relative humidity on the slenderness limits was studied for two different steel ratios: = 0. including the assumption that the quasi permanent service load. As can be seen the difference is important. The ULS load is given by the following expression: qULS = 1. The results are given in Fig. the lower will be the load considered for the verification of deflections.21 it can be seen that the influence of relative humidity on the slenderness ratio is important with a variation of ±15% for the active deflection and ±10% for total deflection from the reference value of 70% of relative humidity.35 (g1 + g2) + 1.5% reinforcement ratio and l/d=14 for 15% reinforcement ratio. The lower the value of the permanent load.5%.5. Table 7.3.3. The figure also shows that the influence of relative humidity does not significantly depend on the steel reinforcement ratio. 70 and 80%.21.Guide to EC2 Section 7 7.4. g2/qtot and q/qtot are given. Figure 7. In Table 7. 7.
fl) The final version of prENV. allows the use of the flexural tensile strength of concrete instead of the mean tensile strength for the calculation of deflections. This topic was subject of much controversy and it is therefore interesting to test the differences to which this may lead. agrees well with the values of prEN 19921 table 4. much higher slenderness limits can be established. 7. Fig.23. These curves form the basis for the formula proposed for the slenderness limit in prEN 19921. For this comparison. This value.11 Influence of the concrete grade The influence of the concrete grade on the slenderness limit has been studied in a very detailed manner for many values of the reinforcement ratio. These curves.14 which are also in accordance with the values of MC90 and ENV199211:1991.4. which is dependent on the concrete grade.3. The curves also show that for low steel ratios.23 shows the results of this comparison.Guide to EC2 Section 7 Comparison Different load distribution (g1/qtot). (g2/qtot). Page 720 Table of contents . The concrete grades considered are: C30.C60 and C100. as explained in section 7.4. 7. (q/qtot) Figure 7.3. but. C40.12 Influence of the tensile strength of concrete (fctm vs.10. fctm. the relationship between total service load and ultimate load has been taken as 51%. Slenderness limits for total deflection and different total service load to ultimate load ratios 7.3. and provides a continuous estimation of the slenderness limit as a function of the reinforcement ratio. For this purpose. is different from the reference value. the slenderness limits were redetermined with the assumption that the flexural tensile stress of concrete could be used. It can be seen that there is a very significant influence of the concrete grade on the slenderness limit.22.24.4. Figure 7. The results are shown in Fig. There is a clear discontinuity in the curves which separates the point in which cracking does not occur from the point in which the bending moment is greater than the cracking moment. Slenderness ratio as a function of the concrete grade 7. also show very well the behaviour for very low steel ratios.
4.6%.Guide to EC2 Section 7 Figure 7. As an alternative to this procedure a simplified method consisting in calculating an equivalent moment of inertia as explained in section 2.4 Simplified formulae 7.4.24. For reinforcement ratios larger than 0. Additionally. since cross sections near the point of zero bending moment will not crack.1 Simplified formula of EC2 The application of the general method of prEN 19921 is very tedious and time consuming.t 1 . Slenderness Limit using fctm and fctm.fl As can be seen there is an important difference only for very small reinforcement ratios. This creep coefficient which allows to take into account the load history in a simplified manner is defined as: Ieq = g1I t. since calculations have to be made for many sections. This procedure in on the safe side. the difference is negligible.4. 7. in this comparison and in order to keep the simplified formula simple. an equivalent creep coefficient is used.1 for the centre span of the beam only and assuming this value for the whole beam is also recommended in the present draft for prEN 19921.
+g2 I t.t 2 .
+\ 02 qI t.t 3 .
2.25.67) The general and simplified procedure are compared for the reference values described in section 7.5% reinforcement ratio. Fig. The comparison is made in terms of the differences in the value of the slenderness limit for different reinforcement ratios.25.7. As can be seen. the error is about 5% on the safe side. in Fig. both procedures yield practically equivalent for steel ratios greater than 1%. Figure 7. In this case a maximum difference of 20% in obtained for the lowest reinforcement ratio considered. 7. g1 +g2 +\ 02 q (7.3. For a 0.26 shows a comparison of the differences obtained by plotting the ratio of (L/d)simplified to (L/d)general as a function of the steel ratio.4. Comparison of general and simplified methods of prEN 19921 proposal Page 721 Table of contents .
Comparison of general and simplified methods of prEN 19921 proposal – Evaluation of the relative error 7. which is especially helpful for slab elements.4. Concrete Grade C30 Page 722 Table of contents .23. and in order to provide designers with a continuous relationship between the slenderness ratio and the reinforcement ratio. In Fig. which is important. especially for low steel ratios.11.70) These formulas are a mathematical approximation to Fig.27. and mostly on the safe side.Guide to EC2 Section 7 Figure 7. this approximation is evaluated. the following expressions have been proposed.' © ¹¼ ¬ if < 0 (7.5 fck +3.5 fck 0 +3.3.5 Formula proposed for slenderness limit In order to take into account the influence of the concrete grade on deflections. 7.68) 0 = fck 10 3 § 0 · 2º l ª 0 =k «11+1. As can be seen the degree of coincidence between the formula and the calculated values of the slenderness limits is very good. Four plots are presented for the four concrete grades considered in section 7.26.2 fck ¨ 0 ¸ » ¨ ¸» d « . 7.2 fck ¨ 1¸ » d « © ¹ » ¬ ¼ § ·º l ª =k «11+1. 3 (7.4.69) if t 0 (7.
Comparison of General procedure with simplified formula for slenderness limit Page 723 Table of contents .Concrete Grade C45 Concrete Grade C60 Concrete Grade C100 Figure 7.Guide to EC2 Section 7 Formula .27.
Zur Frage der Rissbildung durch Eigen – und Zwangspannungen in Folge Temperatur im Stahlbetonbau. May 1993. Serviceabillity Limit States. Falkner. Charif. Favre. Publication Nº 114 (1986). 4. Rüsch. Cracking and Deformations. Jaccoud. R. Eurocode 2: Design of concrete structures – Part 1: General rules and rules for buildings. H. H. CEB Manual (1985). Rüsch.3 Deformation. Beeby. Ref. Versuche mit Betonformstählen Teil III. July 1999 [3] CEBFIP. Mallée. M. prEN 19921:2000. [2] Beeby. Model Code 90. Draft of chapter 15. TH Damstadt (1985). [3] CEB. Die Arbeitslinie eingebetteter St¨¨ale bei Erst – und Kurzzeitbelastung. Deutscher Ausschuss Für Stahlbeton Heft 208 (1969). Rehm. [4] CEN. Rissenbreitenbeschränkung im Stahlbeton und Spannbeton.4 (2001). Final Draft (2001). Rehm. [9] M. Versuche mit Betonformstählen Teil II. No.3. Bulletin d’Information Nº213/214. fib Bulletin 2 – Structural concrete Textbook. Beeby. Untersuchungsbericht (1976). Versuche mit Betonformstählen Teil I. [4] EC2. Elighausen.W..und Spannbetontragwerken. 1st Draft. prEN 19921:1991. 19921. Ref. Rissverhalten von Stahlbetonkörpern bei Zugbeanspruchung. [10] PrEN. Krips. [13] G. [5] R. [8] J. Rüsch. [2] A. A. H. Deutscher Ausschuss für Stahlbeton Heft 165 (1964). J. Rehm.. Teil 11 (1996). Koprna.P. Planung von Stahlbeton.W. Deflection [1] CEN. Page 724 Table of contents . Calculation of Crack Width. Volume 2. [12] G. Deutscher Ausschuss für Stahlbeton Heft 160 (1963). [6] M. Rehm. H. [7] G. [11] G. Armature Minimale pour le Contrôle de la Fissuration. (1987). Model Code 1990. Jaccoud.3. PrEN 19921 (Final draft) Chapter 7. Eurocode 2: Design of concrete structures – Part 1: General rules and rules for buildings.Guide to EC2 Section 7 REFERENCES Cracking [1] A. Deutscher Ausschuss für Stahlbeton Heft 140 (1963).P. Dissertation (1977).W. G. No. R. Hartl.
Basic value of factor b for transmission length according to ENV.5. However. table 4.8 Additional rules for large diameter bars 8. For anchorage at the ultimate limit state.10. The basic value of the transmission length for prestress at release (transfer) is given as a multiplier b for the nominal diameter.9. 8.6 were mainly editorial. but nothing is said about reducing the transmission length for lower values of prestress A ”neutralized zone” for the case of sudden release is mentioned in 4.1 General 8. The national comments on clause 4. given in ENV 199213 should also be mentioned. The ENV rules did not take into account the following parameters: a) the initial prestress in the tendons1 b) the way of release of prestress (gradual or sudden)2 c) the difference in bond conditions with regard to the position of tendons etc.2.1.2 times the basic value.1 Summary of rules The rules in MC90 and EN 199211 are summarized in table 8. Background The ENV rules for the anchorage of pretensioned tendons are found in 4. without details. Table 8. DETAILING OF REINFORCEMENT AND PRESTRESSING TENDONS .9 Bundled bars 8.10.2 of the EN are basically a different formulation of clause 6.8 = 1.2/0.15 See example 6.4 Anchorage of longitudinal reinforcement 8.10 Prestressing tendons SECTION 8. an opening for “alternative buildup of prestress” is given.5.2. compared to MC90.6 (5). like in the ENV.15 See example 6. They are described and commented more in detail in the following clauses.2.2. whereas ”pullout” refers to anchorage in ULS Table of contents Page 81 .GENERAL See example 6. of which the more unfavourable one should be used depending on the design situation. with a ratio 2.2 Spacing of bars 8.7 Two design values of the transmission length are defined. but with a ratio 1. which represents a more uptodate knowledge. the bond is assumed to be the same as at release of prestress. but the rules needed a thorough update based on MC90. however. and takes into account all the important parameters. This has not been explicitly included in the EN.5 Anchorage of links and shear reinforcement 8.7 (table 1 above) is valid for the maximum allowable prestress.6 Anchorage by welded bars 8.5.11 in MC90.0 between the upper and lower value. 1 2 3 It is stated in the ENV that table 4.3. table 8. the upper value will be o lower and the lower value somewhat higher according to the EN. The rules in 8. since a linear development is considered to be more realistic. but no value is given ”Pushin” refers to the situation at release of prestress.2.Guide to EC2 Section 8 SECTION 8 DETAILING OF REINFORCEMENT AND PRESTRESSING TENDONS GENERAL 8.15 C8.10.2.2 have been calibrated to give the same mean value as MC90.2.3.1.8 and 1. 0.1. MC90 gives two design values of the transmission length.2. The present rules in clause 8.7 Laps and mechanical couplers 8.10. There is one difference.2. Thus. d) the difference in bond conditions with regard to “pushin” and “pullout”3 e) the difference between strands and indented wires.10.3 Permissible mandrel diameters for bent bars 8. 8. Rules in MC90 and EN 199211 8.5 between the upper and lower values.3.2. Among the ENV rules the alternative parabolic development of prestress.6.10.
2 for 7wire strands Kp2 takes into account the position of the tendon: Kp2 = 1.2.2.10.2.10. fctd(t) is the lower design value of concrete tensile strength at time of release. 8.2.1 Basic anchorage length lbp = A sp fpd SI fbpd where Asp/SI = I/4 for circular tendons.10. Kp2 = 0.2.2. 8.9. Summary of rules in MC90. see below 8.2. (7/36)I for 7wire strands fpd = fptk/Js design value of tensile strength of tendons fbpd design value of bond strength.2.2 Rules in MC90 8.2.7 for all other cases. 6.4 for indented and crimped wires and Kp1 = 1.2.0 for good bond condition.2 8.25 for sudden release considers the action effect to be verified: Table of contents .2 Design value of bond strength fbpd = Kp1Kp2fctd(t) where Kp1 takes into account the type of prestressing tendon: Kp1 = 1.3 Transmission length lbpt = D 8 D 9 D10lbp Vpi fpd where D8 D9 Page 82 considers the way of release: D8 = 1. or at 28 days for verifications in ULS.0 for gradual and 1.11.Guide to EC2 Section 8 Table 8.2.2. and EN 1992.10.10.
7 otherwise.mean which gives Kp1.10.1.3. bending moment etc.10.1 Bond strength The bond strength governing the transmission at release of prestress is fbpt = Kp1 K1 fctd(t) where Kp1 takes into account the type of tendon and the bond situation at release (”pushin”) = 2.5 × 0.3. The “basic anchorage length” used as a separate parameter in 8.10.EN = 1. comparable values of D9 would be 0.7 for indented wires (“crimped” wires are hardly used anymore. 8.2.19) stress in tendon just after release (same as MC90) pi Normally a short transmission length gives higher transverse stresses in the concrete at release (spalling.2. not a lower limit as in MC90 (the upper limit corresponds to D9 = 0.1. The factor p1 here includes two factors.0.1) or = 0.2. even if the criteria in 8.2.2.2 is instead incorporated directly in the expressions for transmission length and anchorage length.Guide to EC2 Section 8 D10 pi D9 = 1.0.5 for the transmission length). the more unfavourable of the following two values should be used as a design value.3 Rules in EN 199211 The rules in 8.0 for good bond conditions (see 8.2 for 7wire strands K1 = 1.2. This is the “bond situation” to which MC90 refers in the definition of D10: when the stress decreases at release there is transverse expansion of the tendons.Vpcs fpd 8. 0. giving a kind of “wedge” effect. splitting and bursting stresses in the terminology of MC90).10.5 and 1.3.7 for indented or crimped wires is the steel stress just after release Vpd .10. 4. namely the factor D10 and the mean value of D9.2 are the same as in the ENV. Therefore.2 for upper and lower design values.8 and 1.2.2.2.EN = Kp1.1.05(t) /JC. fbpt is here a mean value of the bond strength. design value of tensile strength at the time of release.3.4.7 0.1 are not met.5. However. which in MC90 are applied to the transmission length instead. unless good bond conditions can be verified fctd(t) = fctk.1 The factor K1 is the same as in 8.4.0 for calculation of anchorage length when moment and shear capacity in ULS is considered.7 × 0.6 (4). (0.33.10. The corresponding factor in MC90 is D9 with the values 0.2.25 for circular tendons. 8.2 Transmission length The basic value of the transmission length is lpt = D1D2 I pi/fbpt where D1 = 1.1 Transfer of prestress 8.1.25 for sudden release (same as D8 in MC90) D2 = 0. related to the compressive strength at the same time according to table 3.1 (and the same as p2 in MC90). and D9 = 0.MC90 D10 D 9.10. Page 83 Table of contents .2 = 3.3.0 for gradual release.4 = 2.EN = Kp1.10.2.5 for verification of transverse stresses in the anchorage zone considers the influence of bond situation: D10 = 0.75 for wires for strands Thus. depending on the design situation: lpt1 = 0.) = 3. has been added for cases where good conditions can be achieved by other means. fbpt includes the favourable effect of “pushin” at release.2 0.5 + 1.4 Anchorage length in ULS lbpd = lbpt + lbp where pd pcs tendon stress under design load ( pd fpd) tendon stress due to prestress including all losses 8. 1.8 lpt lpt2 = 1. and verified.2. MC90: 7/36 § 0.3.75: Kp1. Furthermore. whereas a long transmission length is more critical for ULS with regard to shear.2.2 lpt The factors 0.2. a possibility to assume “good bond conditions”. see 8.4.67 and 1. therefore not incl. there is an uncertainty in the calculated value. Furthermore. With the mean value of this factor included elsewhere. See 8.2.75 1.10.5 for strands and D10 = 0.2 are summarized below and compared to those in MC90.19 for 7wire strands (cf.2.0)/2 = 0.
10.10. and there is no favourable effect of “pushin”.3.1: lbpd = lpt2 + D2 I ( pd .p) / fbpd where lpt2 the upper design value of the transmission length (see 8.2.10.3.3.2 Anchorage length The anchorage length is based on the upper design value of the transmission length and the increase of stress above the remaining prestress.1 applies also to figure 8. For fctd there is a limitation to the value valid for C55.2 for 7wire strands 1 same as above fctd = fctk. This may be taken into account. the ENV is by far on the unsafe side.3. MC90 and ENV is shown in figure 8.2. The reason is that the linear relationship between fbpd and fctd cannot be expected to be valid for higher concrete strengths. The figure covers indented wires and strands. same as in MC90) 8. The straight line representing the ENV is the same in all 6 diagrams.05 / JC. The comment to figure 8.Guide to EC2 Section 8 8.1 Transmission length 8. where the tendons “shrink” in the transverse direction when the stress increases. The EN and MC90. whereas according to the EN and MC90 the transmission length is directly proportional to the prestress.2. design value of concrete tensile strength p2 has the same values as p1 in MC90. but requires a special verification.2.10. nor the way of release (at least does not tell how to do it).10.10. nor does it take into account bond conditions. The most notable difference in this case is that the ENV does not take into account the magnitude of the initial prestress at all (at least it is not explained how to do).3. with bond strength according to 8.1 as a function of the concrete strength.3 the total anchorage length in ULS is shown as a function of the stress to be anchored.3.10.2.2. do take these parameters into account. although not in proportion to the tensile strength.2. Due to this the comparison gives very different results. since the ENV does not make any difference between strands and indented wires.1 Effect of concrete strength The transmission length according to EN.10. particularly if released suddenly.1.3 Anchorage in ULS 8.3. The average bond strength will normally increase also above C55.2.2. Instead the bond situation at anchorage is characterized by “pullout”. for strands released gradually and in good bond conditions.2.10.1. Numerical comparisons 8.3. Page 84 Table of contents .3.2. 8.3.2.0. 8. The lower design value of the transmission length is shown.2. For example.2) pd tendon stress to be anchored (same as in MC90) p the remaining prestress (after all losses.3. which results in a more uneven distribution of the bond stress.1 Bond strength The bond strength for anchorage of stresses above prestress is fbpd = Kp2 K1 fctd where p2 takes into account the type of tendon and the bond situation at anchorage (”pullout”) = 1. This is due to an increasing brittleness.10. on the other hand.3.2 Anchorage length In figure 8.2 Effect of initial prestress Figure 2 shows the transmission length as a function of the initial prestress. whereas for indented wires released suddenly and in bad bond conditions.2. Like for the transmission length. the ENV is conservative for gradually released strands in good bond conditions.3.4 for indented wires = 1.10.2. different bond conditions and the way of release.2. the ENV is extremely conservative compared to the EN and MC90. and far on the unsafe side for wires in bad bond conditions. 8.2.3.1.2.1 in the third paragraph of 8.
Thick black line = EN.1.Transmission length (lower design value lpt1) as a function of the initial prestress pi. thin black line = MC90 Figure 8. Initial prestress is 1000 MPa.Transmission length (lower design value lpt1) as a function of the concrete strength fck(t) at release. Concrete strength at release is 40 MPa.thin black line = MC90 Page 85 Table of contents . Thick black line = EN. thick grey line = ENV.2.Guide to EC2 Section 8 Figure 8. thick grey line = ENV.
pd.Guide to EC2 Section 8 Figure 8. nominal concrete strength is C50 and initial prestress is 1000 MPa. Thick black line = EN. thick grey line = ENV. thin black line = MC90 Page 86 Table of contents . Anchorage length lbpd as a function of the stress to be anchored. Concrete strength at release is 40 MPa.3.
For normal density concrete the sustained loading factor is defined to be 1. The first one describes the average behaviour as realistically as possible and is meant for calculating the distribution of forces and moments in a structure. a StateoftheArt Report giving a good overview of common practice with regard to design practice in various countries [1]. In this respect the joint CEB/FIP (now fib) Task Group 8.5. 11. 11. so that fcd = 0. Important reports produced by the partners in this project were [3] and [4].1.1 published in 1999 fib Bulletin 4 “Lightweight Aggregate Concrete: Codes and Standards”. as alternatives. Altogether 3 stress strain relations have been defined. Another report “Structural Lightweight Aggregate Concrete: Recommended Extensions to Model Code 90”.85 and dividing by a material safety factor of 1.1. Chapter 10 in the new draft for EC2 of 1/1/2000 applies to all concretes with a closed structure made with natural or artificial mineral lightweight aggregates. where the first figure stands for the characteristic cylinder strength and the second for the characteristic cube strength. unless reliable experience indicates that provisions different from those given can be adopted safely. The second and the third differ only in the ascending branch.57fck.1 General SECTION 11 LIGHTWEIGHT CONCRETE C11. Other valuable work was done within the scope of the BriteEuram Project “EuroLightcon”. the second being parabolic and the third being linear.3 Material properties The material properties of lightweight aggregate concrete are related to the corresponding properties of normal aggregate concrete as defined in section 3.1 General A favourable circumstance for the preparation of Chapter 11 was that during the last years substantial work was conducted in updating the stateontheart on lightweight aggregate concrete to the most actual level. Lightweight concrete is more brittle than normal density concrete of the same strength class. Fig. The following conversion factors have been introduced in order to derive the properties of lightweight concrete from those of normal weight concrete: KE K1 K2 K3 U conversion factor for the calculation of the modulus of elasticity coefficient for the determination of the tensile strength coefficient for the determination of the creep coefficient coefficient for the determination of the drying shrinkage ovendry density of lightweight aggregate concrete in kg/m3 Concrete strength and stressstrain relations The strength classes for lightweight aggregate concrete range from LC 12/15 to LC 80/95. Figure 11. by the same Task Group. Design stressstrain relations for concrete in compression The maximum stress in the diagram is obtained by multiplying the characteristic cylinder strength with a sustained loading factor of 0. The second and the third are both given. This is reflected in the formulation of the ultimate strain.3 Materials C11.Guide to EC2 Section 11 SECTION 11 LIGHTWEIGHT CONCRETE 11. “Lightweight aggregate concrete” is defined as a concrete having a closed structure and an ovendry density of not more than 2200 kg/m3 consisting of or containing a proportion of artificial or natural lightweight aggregates having a particle density of less than 2000 kg/m3. for the design of cross sections.1. where the factor K1 has been added. is now ready to be published [2]. unless Page 111 Table of contents .
the Emodulus is calculated from Ecm = 9.7 fctm Emodulus for concretes <C50/60 (11.3 for lightweight concrete classes lower than LC20/25. will occur after a considerable time. Similar results were obtained by Smeplass [6].3(U/2300)2 can be used.5(1P)).1. The sustained loading effect is therefore with high probability compensated by the gain in strength between 28 days and actual loading of the structural member (see further the background report of chapter 3. Therefore in 10.5 Dlcc should preferably be defined as 0. Weigler [5] reported that the strength of lightweight concrete under sustained loading was only about 7075% of the short term strength.2a) (11. Here it should be noted that the average tensile strength fctm normal density concrete follows from fctm = 0. For intermediate values of U linear interpolation may be applied. if occurring anyhow.5) In the the old version ENV199214 it is denoted that for lightweight concrete the creep coefficient I can be assumed equal to the value of normal density concrete multiplied by a factor (U/2300)2 for U > 1800 kg/m3. the stresses in the paste will remain higher and so the creep of LWAC [4].40 + 0. Kordina [5] states that creep is a matter of the cement paste and not of the aggregate. The tensile strength of lightweight concrete can be obtained by multiplying the corresponding strength of normal density concrete of the same strength class with a factor K1 = 0. Since most of the lightweight aggregates have a lower stiffness. Since further research seems to be necessary here.12 ln (1+fcm/10) for concretes > C50/60 the characteristic (5%) value follows from fctk = 0. the more load will be carried by the aggregate skeleton and the more the stresses in the paste will decrease. For normal density concrete.1) where U is the upper limit of the ovendry density. according to chapter 3. the Japanese Code JSCE and the German code DIN4219 give formulations with the same tendency.2b) (11. So when loaded the concrete will be much older (maybe even years) than 28 days.4) In the previous part of the Eurocode. according to [1] also other codes like the Norwegian Code NS 3473. this would support the idea of introducing a sustained loading factor for lightweight concrete anyhow. The results are explained by creep of the matrix. in spite of the fact that. The stiffer the aggregate. however. Furthermore the creep strain has to be multiplied by a factor K2 = 1.3. Anyhow. There is however serious doubt on the correctness of the statement in ENV 199214 that the creep of lightweight concrete is smaller than that of normal density concrete.60U/2200 (11.1. Neville [6] developed a twophase model where he distinghuishes the cement paste and the aggregates as two parallel load bearing components. which would imply that similar compositions of LWAC and NDC should give the same specific creep.1). Furthermore it is reported that the sustained loading effect is more pronounced than in normal aggregate concrete.Guide to EC2 Section 11 specified otherwise (Clause 3.85 “unless specified otherwise”. For U < 1500 kg/m3 a factor 1.3) An estimate of the mean value of the secant modulus Elcm for LWAC can be obtained by multiplying the corresponding value for normal density concrete by the coefficient KE = (U/2200)2 (11. existing information seems to confirm that there is not difference between normal density concrete and lightweight concrete with regard to the Page 112 Table of contents . ENV 199214 “General rules for lightweight concrete with a closed structure.5(fck+8)1/3 Creep (11. overloading the aggregates. Consequently this phenomenon occurs when the strength of the aggregates it utilized to its maximum [4]. The most important reason is that sustained loading effects. the equation KE = (U/2200)2 was mentioned. In lightweight concrete.30 fctk2/3 and fctm = 2. the increase in strength after 28 days is smaller than in normal weight concrete [4]. A decrease of the stresses in the paste will result in smaller creep deformation of the paste and hence of the concrete.
see f. Autogenous shrinkage is believed to be caused by “selfdissication”. This volume reduction is sometimes also denoted with the term “chemical shrinkage”. In [8] it is reported that the final shrinkage of LWAC is about 11.2 [7]. The best formulation seems to be that creep of LWAC is the same as creep of NDC and can be calculated with the same formula’s. 20 specific creep [105 N/mm2] NDC LWAC 15 10 5 0 10 20 30 40 50 60 70 strength at time of loading [N/mm2] Figure 11. Fig. values which were about 35% higher than for NDC. A new element in the formulation of shrinkage is the component Hca which represents autogenous shrinkage.55% in axial direction and 0. Theissing [10] reported.2000.5 times the final shrinkage of NDC of the same strength.2. and goes along with a decrease of the relative humidity in the pore system. Probst [11] reported about shrinkage tests on three different LWAC’s made with Liapor (fcc = 65 Mpa). chemical shrinkage. which is a result of a volume reduction of the hydration product compared to the volume of the reacting water and cement. Hoffman and Stöckl [9] reported for LWAC’s with cylinder strengths of 4050 MPa differences of about 30% with NDC.f where Page 113 Table of contents (11. on the basis of a literature survey.f(t) = Ecc(t) Hca. values of 0.7a) . for concretes with a cylinder strength of 21 Mpa.Guide to EC2 Section 11 specified creep.e. [4]. This is about the same as found for NDC.5 LC20/25 and higher K3 = 1.85 % in transverse direction. with low water/cement ratio’s.2 Since the information from literature is not fully consistent and the values given in ENV 199214 will presumably be not be too far from reality they have been maintained in the new draft of 1. representing the drying shrinkage strain.1. Final specific creep as a function of strength at the age of loading [7] Shrinkage For normal strength concrete the shrinkage is formulated as Hcs = Hcd + Hca where Hcs Hcd Hca final shrinkage strain drying shrinkage strain autogenous shrinkage (11. as given in ENV 199214 and provisionally adopted in the version of EC2 of 1/1/2000 seems to be necessary. i. kept under a RH of 65%. 11.6) The component Hcd. Berwilit (fcc = 43 Mpa) and Leca (fcc = 33 Mpa). is known to be higher for lightweight concrete.i. For normal density concrete the component of autogenous shrinkage is formulated as: Hca. The attention to this additional type of shrinkage contribution was drawn during the introduction of high strength normal density concretes. This drop in relative humidity is accompanied by a volume reduction of the matrix. In ENV199214 it is stated that final drying shrinkage values for lightweight concrete can be obtained by multiplying the values for normal density concrete with a factor K3 defined by LC12/15 to LC/20: K3 = 1. A reconsideration of the formulation for creep of LWAC.
the shear capacity of reinforced (normal density) concrete members without shear reinforcement has been formulated as: VlRdc = [0. Taylor/Brewer [15]. the expression has been verified with 86 test results. as described in EC1 Basis of Design is suitable.18/Jc) in order to show explicitly the safety margin. from Ivey/Buth [12]. Ultimate bearing capacity of LWAC structures With regard to the bearing capacity of structures in the ultimate limit state specially the behaviour in shear and punching is important.162 with a standard deviation of 0. For such a case the classical level2 method.12 has been replaced (0.5 (fck – 10) 103 (11. which have generally a much lower strength than gravel aggregate particles. Walraven [13]. corresponding to a coefficient of variation of 0. (11. which is particularly important since in the draft of 1.02 average longitudinal prestress in the cross section conversion term from NDC to LWAC. A design value can be obtained on the basis of a statistical evaluation.Guide to EC2 Section 11 Hca. Torenfeld/Drangsholt [17]. Therefore for LWAC the contribution of autogenous shrinkage as given by Eq.2000 the standard method has not been involved anymore and only the variable inclination method is given as the basis for the calculation of the shear reinforcement.7b) and Ecc(t) is the hardening function. The contribution of autogenous shrinkage decreases considerably with increasing strength of the concrete. The way how to deal with this method has been described and illustrated by Taerwe [20]. In that case the possible supply of water from the aggregate to the drying microstructure will prevent a significant drop of the relative humidity in the paste and will thus reduce autogenous shrinkage.145. Page 114 Table of contents .3 shows the results. A mean value of vtest/kK1(Ufcm)1/3 of x = 0. In lightweight aggregate concrete the conditions are quite different if the aggregate particles are saturated with water. This difference might also limit the redistribution capacity of the concrete web (rotation of compression struts to lower angles).12K1k(100Uflck)1/3 – 0.f = 2. Evans/Dongre [16]. On order to seen if this formulation is also suitable for lightweight concrete. see Eq. Those questions will be systematically treated. Shear capacity of reinforced concrete members without shear reinforcement In Section 6.15Vcd] bwd where flck k Ul Vcd K1 characteristic cylinder strength of lightweight concrete size factor = 1 + (200/d)1/3 2.0235 is obtained.0 longitudinal reinforcement ratio = Asl /bwd < 0. 11. whereas in lightweight concrete the crack intersects the aggregate particles. Fig. (EC3.8) The coëfficient 0. Hansson [14]. This holds particularly true because cracks in lightweight concretes are supposed to be smoother than cracks in normal density concrete: in normal strength concretes of moderate strength cracks are propagating around the aggregate particles.1.1. 11. Thorenfeld/Stemland [18] and Aster/Koch [29]. This might reduce the shear friction capacity of the cracks and as such reduce the total shear carrying capacity.10/11) has to be regarded as an upper value.
sy zf ywd cot T s (11.20 0. with vertical shear reinforcement.06 0. a reliable design equation can be derived from test results with the general formulation BRd = PBR(1 . the shear capacity is the smaller value of A sw (11.15Vcd] bwd This agrees with the proposal given in [2].22 0. Drangshold [17] Thorenfeld.28 0.This would then result in a coefficient 0.9) with PBR = 0. DBR = 0. Shear capacity of members with shear reinforcement In the new version for EC2. z = inner lever arm of the crosssection.10 K1 k (100U fck)1/3 – 0. 11.DBr E GBR) where BRd PBR DBR E GBR design value mean value of tests sensitivity factor for BR normally taken 0.091 is obtained.8 in the case of one dominating parameter target safety index. and VRd.095 for a concrete class of LC80/95. In this derivation. This would mean an increase of the coefficient 0.02 0 Ivey.max = bw z Q fcd / (cotT + tanT ) Page 115 (11.10 0.Guide to EC2 vu / K1 k (100Ufcm) Section 11 1/3 0.24 0. according to the Model Code.091 with 9% for C25 and 3% for LC 80/95. Verification of Eq.14 0. the mean concrete cylinder compressive strength has been used. only one method for the design of members with shear reinforcement is given.12a) VRd .08 0. s = stirrup distance. whereas in the code expression the 5%lower value fck is used.055 for a concrete LC 80/95.145 a value for the design coefficient in Eq. Stemland [18] 0 10 20 30 40 50 60 70 80 fcm Figure 11. however.3.10) is used.100 for a concrete class LC25/30 and 0. Dongre [16] Thorenfeld. 11. taken 3. For members of normal density concrete not subjected to axial forces. contrary to ENV 199211. fywd = design yield stress of shear reinforcement.15 for a concrete LC 25/30 and Gf = 0. 11. T = inclination of compression strut.8 should be modified to V1Rdc = [0.26 0.162. This method is based on the variable angle truss model.18 0. The conclusion is that Eq.12b) Table of contents .11) where Asw = crosssectional area of one stirrup. the relation fck = fcm – 8 (Mpa) (11.8 coefficient of variation (11.8 for the shear capacity of members without shear reinforcement with test results According to the level 2 method. This means coefficients of variation Gf = 0.8 of 0.16 0.12 0.8 and GBR = 0. Brewer [15] Evans. Buth [12] Walraven [13] Hanson [14] Taylor.04 0. In the new version of EC2.
as a function of the shear force for gravel concrete (left) and lightweight concrete (Aardelite. medium (M) and low (L) shear reinforcement ratio’s [21] On the basis of those observations. with low (GD30L). The tests show that the rotational behaviour of the inclined struts is similar for LWAC and NDC.Guide to EC2 Section 11 With the additional condition A sw f ywd d 0. for high (H). In normal density concrete during crack formation the strong aggregate particles do not fracture and the crack propagates around them: therefore the crack surface is very rough so that large frictional forces can be transmitted. both for the course and the sand fractions. so that the final rotation remained relatively small. but this was compensated by the roughness on the macrolevel. 1975 kg/m3 (Lytag) and 1780 kg/m3 (Liapor). The inclination of the concrete struts can freely be chosen between 21. Fig.4 shows two diagrams. The left diagram shows the results for the gravel concrete members. On a mesolevel the roughness of the cracks in lightweight concrete is indeed smaller. or if a higher lower limit should be defined. medium (GD30M) and high (GD30H) shear reinforcement ratio’s. In order to answer this question tests have been carried out on Ishaped beams with varying shear reinforcement. In lightweight aggregate concrete the aggregate particles are intersected. Fig. 11.and LWACbeams behave similarly can be explained by the overall shape of the cracks. for the design of members with shear reinforcement in lightweight concrete.4. with sufficient capacity to develop the necessary transmission of forces across the inclined cracks. whereas furthermore the beams were exactly similar.4.12c) The first eq.5) and 450 (cot T = 1). the same principle as for normal density concrete was maintained in the new version for EC2. so that a less rough crack surface is obtained. Any series consisted of three beams.8° (cot T = 2. 50 45 40 35 30 25 20 0 100 200 300 400 L H M T gravel 50 45 40 35 30 25 T lightweight M H L 20 500 600 0 100 200 300 400 500 600 V [kN] V [kN] Figure 11.5% he found a lowest strut inclination at failure of 250. the right diagram shows the corresponding curves for lightweight concrete. Liapor and Aardelite. In this way also in the interface contact areas occurred. On the web the state of deformation was continuously measured.5). Principal strain directions T in relation to the longitudinal member axis.5vf cd bws (11. In those concretes only the coarse aggregate particles were of the lightweight type: the mixtures contained natural sand. Walraven [21]. right).5 (11. It is therefore questionable whether the rotation capacity of the web is sufficient to allow as well a lowest strut inclination of 21.12b crushing of the inclined concrete struts. in which the inclination of the principal strain is represented. caused by the overall crack undulation.12a represents yielding of the shear reinforcement and the second equation 11.8° as a minimum.8°. In both cases the beams with the lowest shear reinforcement ratio showed the highest strut rotation capacity. Q is an efficiency factor for the concrete crushing strength depending on the concrete strength according to: Q = 0. The unexpected result that NDC. Page 116 Table of contents . so that the inclination of the principal compression strain could be monitored. The concrete volume weights were 2050 kg/m3 (Aardelite). This is a very important condition to allow a rotation of the inclined struts from 450 down to an angle of 21. which contained whether low. including a lower limit for the strut inclination of 21. medium or high ratio’s of shear reinforcement. Obviously the other two shear reinforcement ratio’s were both too high to reach yielding of the steel.13) An important question with regard to the applicability of those formulations for lightweight aggregate concrete is if the concrete struts in the web have a sufficient capacity to rotate.80 (cot T = 2. Also Thorenfeld [22] reported a substantial decrease of the strut inclination with increasing load. His tests were carried out on lightweight concrete with Leca aggregates. Three types of lightweight aggregates were used in the various concrete mixes: Lytag. The volume weight of this concrete was 1500 kg/m3. For a shear reinforcement ratio of 0. 11. 11.6(1 – fck/250) 0. Those series were compared with a reference series with beams made of normal density concrete.
12.12ac and 11.17 (only reinforced concrete. Although the number of available tests was limited. Implicitly this means that a reduction of the efficiency factor Q is necessary.c = 0.12 is correct. was formulated as Q = 0.12/15 for lighweight concrete members with shear reinforcement by test results 0. 11. Comparison with tests shows that the combination of the equations 11.12 1k (100 Ul fck)1/3 – 0. which means a coefficient of variation of 0.50(1 – fck/250) Punching shear capacity of lightweight concrete slabs and column bases.2 0.17) (11. Hamadi [22] and Thorenfeld [23].6 shows a comparison between Eq.181 – 0.70 0. the efficiency factor Q.08 Vcd (11. A better formulations is therefore QLWAC = 0. Hognestad [26].15a) Fig.6 (1 – fck/250) 0. 0.425 K1 (11.80 0.10 should be applied in stead of the factor 0.10 0 Walraven [20] Hamadi [21] Thorenfeld [18] vu / vfcm 0 Ufsw / vfcm Figure 11. The level2 method yields then a design value.5 shows the comparison of the combination of Eq.17 is the same as the equation used for shear. Regan [25].0183 = 0. Corley [27] and Ivey [28].16 can be maintained (for the 22 tests a mean value of PRd = vtest/vcalc =0.5 0.14 could be simplified to QLWAC = 0.1 0.8 3.15 with test results from Walraven [21]. so Vcd = 0) and test results by Tomaszewicz [24]. Fig.8 0.85 K1 QNDC or QLWAC = 0. reducing the allowable inclined compressive stress.3 0. 11.85 K1 0.10.16) which is multiplied with a basic control section at a distance of 2d from the loaded area.9. it may be wondered whether also here. 11. An analysis showed that this can not be solved by restricting the strut rotation to a higher value of Tmin. In normal density concrete the punching shear capacity is calculated with the nominal design punching shear stress vRd.9 Eq. a factor 0.5. 11.ac and 11.8 0. Page 117 Table of contents .125).14 does not give an appropriate lower bound.Guide to EC2 Section 11 In ENV 199214.6 0.4 0. Since Eq.12 basically valid for NDC. a statistical derivation according to the level 2 method leads to the conclusion that the coefficient 0. of xd = 0.7 0.60 0.50 0. 11.40 0. 11.181 is obtained. including the model uncertainty. so that Eq.11. defining the crushing capacity of the concrete struts. like in the case of Eq.30 0.0183.425 (11.14) The efficiency factor Q is only slightly smaller than the corresponding expression for NDC in the new EC2 version. the part on lightweight aggregate concrete.15b) (11. with a standard deviation of 0.20 0. Verification of the design equation 11. 11.6 – flck/235 0.
because of the minimum microcracking.6. 11.8).7.0 0 10 20 30 40 50 60 70 fcm [MPa] Figure 11.7.Guide to EC2 vu. Efficiency of a confining reinforcement in different concretes according to MC’90. as treated in [1] and [2]. Finally.05 0. 11. although Poissons ratio in the elastic region is almost the same as for NDC. Verification of Eq. the transversal strain of LWAC at the maximum compressive load is in part considerably lower.15 0. strength under multiaxial state of stress. Hence.10 Tomaszewicz [24] Regan [25] Hognestad [26] Corley [27] Ivey [28] 0. First.20 0.16 for the punching shear capacity of slabs with test results modification Multiaxial states of stress A short review of the major points on this subject. Fig. This leads to a higher compressibility under multiaxial loads due to the porous nature of LWA. This is valid in particular for LWAC with lightweight sand. the lower ratio between the tensile and the compressive strengths reduces the compressive bearing capacity of locally loaded LWAC (Fig.exp / K1 k (100Ufcm) 0. fib Bulletin 8 and CEB Bulletin 228 Page 118 Table of contents . 11. is given by Faust in [30]. local compression and efficiency of confinement. Figure 11. The second reson is the lesser resistance of LWA to lateral pressure in comparison to dense aggregates. The text is cited here: Major differences between LWAC and NDC arise from differences in transversal behaviour. increasing effects of lateral pressure on confined concrete sections are generally reduced in LWAC.25 1/3 Section 11 0. All of these aspects are connected with each other and attributed to various LWAC phenomena.
[31] Deflection of slabs Since the Emodulus for LWAC is smaller than for NWAC this will influence the acceptable limits for l/d. where KE is defined as Vs E s (d x ) G 5 48 Vs l 2 § U · (11. 11.19) d Vs For the height x of the compression zone in the linear elastic cracked state it is known that x nU1 (nU1 ) 2 2nU1 (11.2 in prEN199211 © 2200 ¹ In Table 11. Local compression in LWAC according to Walraven et al.20) d where U1 = longitudinal reinforcement ratio and n = Es/Ec For a concrete strength C20/25 the Emodulus. see Eq.000/9000 = 22. In general the requirement 0. as a function of U1(%) and KE KE 2 Page 119 Table of contents .8.18) x E s d(1 ) d where Vs is the stress in the longitudinal steel and x is the height of the compression zone.004 l will be governing.2 For lightweight concrete Elc= KE Ec. can be assumed to be Ec # 9000 MPa.19).1) it is found that x 0. so that n # 200.1 the values for x/d are calculated on the basis of Eq. and subjected to a uniformly distributed load is (assuming the cracked state) 5 ql 4 G 384 ( EI) r with Mmax = 1/8 ql2 this can be written as: 5 M max l 2 G 48 (EI) r with M max Hs k r and kr (EI)r dx it is found that (11. (11.Guide to EC2 Section 11 Fig.038E s (1 ) l d (11. spanning one way. Substituting this equation in (11. 11.21) ¨ ¸ . including creep effects. This can be calculated as follows: The deflection of a slab.
217 0.29 0. (11.925 0.89 KE=0.45 1800 0.96 1 0.19 the change of l/d can be calculated.Guide to EC2 Section 11 Table 11.235 0.9 shows the results graphically.2.4 0.34 0.0% 2200 1 0. Table 11.28 0.50%.95 KE=0.30 0.3 0.950 0.31 0.53 0.26 0.89 0. In practical design the reinforcement ratio is generally between 0.972 1 0.974 1 0.1 values for x/d calculated on the basis of Eq.93 0.42 1600 0.0 Reduction factor R for l/d KE=1 KE=0.3% 0.15 and 0.67 0.38 x d 0.19).20 0. 11.93 0.90 0.27 0.2% 0.2 0. 11. Figure 11.34 0.5 1.4% 0. Reduction factor R for l/d as a function of KE and U1 Page 1110 Table of contents . reduction factors for l/d U1 = 0.53 0.29 0.37 2000 0.9.5% 1. It can be seen that the reduction factor K0.82 0.84 Fig.95 1 0.32 0. on the basis of Eq.921 0.2 gives the reduction factors for l/d obtained in this way Table 11.15 covers the calculated E values quite well.24 0.67 0.946 0.82 1 0.26 0. as a function of U1(%) and KE U1 = U(kg/m3) KE x x x x d d d d 0.48 The values x/d being known as (1x/d) is known.
10. No. 9. 208 p. Neville. Paris. E.. CEB Bulletin 219.H.. “Behaviour of High Strength Page 1111 Table of contents . Vol 15. 313.. Stöckl. et al. H. A. Sandefjord. 40 pages 2.. 17.L. Rilem Bulletin. London.S.”Shear capacity of lightweight concrete beams without shear reinforcement”. E. July 1963. Evans.D. 1967. 2024 June 1995. R. J. European Union – Brite Euram III.”Experiments on creep and shrinkage of high strength lightweight concrete”. 244255. Model Unvertainties. 2024 June 1995. Proceedings.. Dilger. AlHussaini.. Second International Symposium. G. Sandefjord. 18. December 1980. 23. Torenfeld. fibTask Group 8. 4. “Lightweight concrete”. No. 5881 (in German). Sandefjord. Thorenfeld.M. Report 5784. pp. L. Xue HY. 44. Dongre. P. 120 (in German). S. Report 2. AlZubi. Walraven.C.”The influence of depth on the shear strength of lightweight concrete beams without shear reinforcement”. pp. 7179. E. 19. pp. Lightweight aggregate concrete – Technology and World Editor G. High Strength Concrete. 8792. Proceedings. 36 pp. 35 pages. Norway. Walraven.3 “Punching Shear Capacity of Reinforced Concrete Slabs”. “Lightweight Aggregate Concrete: Codes and Standards”. Aug.. pp. A. 11. 1993. pp. 561. July 1961.. Tensile strength and diagonal tension resistance of structural lightweight concrete. Berkely 1990. Fisher. J. “Tests on creep and shrinkage of high strength lighweight aggregate concrete”. July 1995. 1978. S. Ramdane KE. fibTask Group 8.”Towards a consistent treatment of model uncertainties in reliability formats for concrete structures”. pp. 1971. Hoffmann.”Structural Lightweight Aggregate Concrete: Recommended Extensions to Model Code 90”. and Tomaszewicz. pp. pp 722. April 1998. Hamadi. pp. Tomaszewicz. 14. Taerwe. Cembureau (1974). Trondheim. Proceedings. Magazine of Concrete Research. December 1998.. Vol. J.. Regan. Lausanne. to be published in 2000.. “Model uncertainties concerning design equations for shear capacity of concrete members without shear reinforcement”. StateoftheArt Report .. 6... 4.. 58. SP2 – Plates and Shells.. STF70 A93082.. Stemland. R. July 1963. 2024 June 1995. ACI Journal. International Symposium on Structural Lightweight Concrete. 1960. the Construction Press.E. CUR report 48. Deutscher Ausschuss für Stahlbeton. European Union – Brite Euram III. Norway. E. Thorenfeld. 13. P. P. “Definitions and International Consensus Report”. N. International Symposium on Structural Lightweight Concrete. Eurolightcon Project. J. Y. 21. A. P. CEBFIP Manual on Lightweight Aggregate Concrete. Ivey. 7. Bologna. K. Kordina.. W. LWAC Material Properties StateoftheArt”.. Brewer. Magazine of Concrete Research. ACI Journal Proceedings..”The suitability of lightweight aggregate (Aglite) for structural concrete”.Drangshold.”Creep of plain and structural concrete”Construction Press. Regan. Probst.pp. London. Norway. 3. Vol 58B. 24. pp. Nr. applications. R. G. 1999.E.1. König.... Deutscher Ausschuss für Stahlbeton... 20. Brooks. Stemland. 91104.. D.. 140. Oct. 5. SINTEF Structures and Concrete.J.. Buth. International Symposium on Structural Lightweight Concrete. Document BE963942/R2. 22.. High –Strength Concrete. 16. H.. pp.”Behaviour of normal and lightweight aggregate beams with shear cracks”. Delft University of Technology.”Shear capacity of lightweight concrete beams”. ACISP121. 9397.V. 25. The Structural Engineer. Hanson. “Experiments on the influence of the mineralogical character of aggregate on the creep of concrete”. A.C. 12..M. “The effects of the type of aggregate on the diagonal cracking of reinforced concrete beams”.. A. Theissing. E.”Shear Capacity of Large IBeams”. 6..”Shear capacity of lightweight concrete beams with shear reinforcement”. CEBBulletin 224.”Shear Capacity of Reinforced High Strength Concrete Beams”.A.1. No. 100. (1993). Stöckl.H. Proceedings.. 1995. Taylor.Guide to EC2 Section 11 REFERENCES 1. Document BE963942/R1. May 1990. Report No. 343. 733744. Nr. 1977 8. 15.
N. Hanson. pp. E.. University of Dundee. Walraven. J. pp. ACIJournal. 11. See example 11.. W. “Recommended extensions to Model Code 90 for lightweight aggregate concrete”..B. Elstner..G. 3. Corley. Shear capacity of lightweight concrete flat slabs”. “Structural Lightweight Concrete: Recent Research”. pp.”Untersuchungen an dicken Stahlbetonplatten”. and Naaktgeboren. pp.A.2 Page 1112 Table of contents ...A. D. den Uijl. Proceedings of International Conference. Scotland. 1974. 26. Cambridge. Buth. 1995. 5 – 30.. Ivey.. Hawkins. J. Aster.. Vol. E&FN Spon. 31. Hognestad. fib news. UK. E. 1. ACI Journal. 29. “Shear strength of reinforced structural lightweight agregate concrete slabs”... 643656. September 79.. 27.M.. June 1964. 153156.C. 1. H. Koch. pp.1. T. October 1968. Shearhead reinforcement for slabs”. 490494. R. Gijsbers. N.L. C. J.Guide to EC2 Section 11 Concrete Slabs”. 2000. M. No.. Faust. Heron. June 1969. 40. Ivy. 811824. No. Universität Stuttgart. R... PP. 30. Concrete 2000. ACIJournal.C. Stroband. 28. 761773. AlZubi. J. J.
This action might not be possible to undo. Are you sure you want to continue?